Here are 9 AI use cases for startups that help you achieve your goals, while improving customer satisfaction, team productivity, and growth.

What we’ll cover:

1. AI in sales: Build stronger customer relationships
2. AI in customer service: Automate support to boost satisfaction
3. AI for marketing: Reach the right audience with targeted campaigns
4. AI in operations: Save time and cut costs
5. AI in finance: Budget smarter and boost security
6. AI in HR: Make hiring easier and boost employee satisfaction
7. AI in security: Keep your data and customers safe
8. AI in product development: Stay ahead of market trends
9. AI in data analysis: Turn insights into action

1. AI in sales: Build stronger customer relationships

Leaders responsible for CRM and AI understand the criticality of data readiness, and in ASEAN, they are rising to meet this moment. AI transforms how startups handle sales, from forecasting to relationship-building. Among AI use cases for startups, sales applications allow teams to build stronger customer connections and predict future needs.

Sales forecasting with AI

AI tools can study past sales data to see trends and predict busy or slow periods. With this information, you can plan better and allocate resources effectively.

• Streamline resource allocation with AI insights.
• Make smarter decisions through clear forecasting.

Sales teams can leverage AI-powered customer relationship management (CRM) like Salesforce to boost efficiency. With tools like Agentforce, your sales team can gain a deeper understanding of customer needs and deliver more relevant recommendations, creating meaningful connections that drive sales. ating lead tracking and allowing reps to focus more on customer relationships.

Improving customer relationships with AI

AI-powered CRM tools store and analyse customer data so you can offer relevant deals and communication. Tasks like data entry are automated, saving your team time.

• Personalise messaging to make customers feel valued.
• Save time by automating tasks and focusing on relationships.

Unlock the Power of Data for AI

See how ASEAN businesses are readying their data for the future of AI. 

Get the guide

2. AI in customer service: Automate support to boost satisfaction

Great customer service can set your startup apart. With AI, you can speed up response times and ensure each customer receives the support they need. This is one of the most widely adopted AI use cases for startups and SMBs, as it enables efficient and scalable customer service.

AI-powered support

AI tools like chatbots and virtual assistants allow you to provide support 24/7. They can answer common questions, solve simple issues, and send complex ones to human agents through collaborative tools like Small Business Slack, creating a smoother experience for customers. In fact, 87% of small business teams use AI to personalise the customer journey across channels, helping enhance customer satisfaction and resolve identity issues.

• Get 24/7 customer support with an AI agent and help customers any time. 
• Resolve customer issues faster with automating routine questions, reducing wait time. 

Reading customer sentiment with AI

AI can analyse feedback from surveys, social media, and customer service interactions to understand customer sentiments better. This helps you fix issues before they grow, keeping customers happy and loyal.

• Build customer loyalty by resolving issues quickly to foster trust.
• Improve product quality by using customer insights to better meet needs.

3. AI in marketing: Reach the right audience with targeted campaigns

AI marketing tools give you insights into your audience, allowing you to create more targeted campaigns that drive higher engagement. AI use cases for startups in marketing are among the most popular due to their ability to increase customer engagement and conversion rates.

Personalised campaigns for higher engagement

With AI, you can use customer data to send highly personalised messages, increasing engagement. With this approach, you’ll see a better return on investment (ROI) and have the flexibility to make real-time adjustments as you track your campaign’s progress. Around 82% of small business marketing teams also leverage AI to offer real-time deals, ensuring customers receive relevant offers when they’re most interested.

• Achieve higher ROI with targeted campaigns that drive more conversions.
• Make real-time adjustments as AI tracks and fine-tunes your messaging.

Predicting trends with AI

AI-powered predictive tools help you spot trends and prepare for customer needs. From product recommendations to ad adjustments, AI lets you stay ahead of the curve.

• Reduce risk by avoiding overstocking or shortages with trend forecasting.
• Stay adaptable by quickly adjusting to meet changing customer needs.

Small & Medium Business Trends Report, 6th Edition

Discover valuable insights from 3,350 leaders of small, medium, and growth businesses (SMBs) worldwide.

Get the report

4. AI in operations: Save time and cut costs

With AI and solutions like SMB Commerce storefront, you can reduce operational hurdles, managing orders and inventory more effectively while cutting down on costs.

Automate routine tasks

AI can handle repetitive tasks like data entry, invoicing, and team workflows. This reduces human errors and lets your team focus on more important work. 88% of SMBs use AI for data integration and process automation, helping reduce repetitive work and improve efficiency.

• Reduce costs with automation for repetitive tasks.
• Ensure clean data with fewer errors for reliable insights.

Optimise store performance with AI

AI can track your storefront performance so you can be more informed and make better decisions: 

• Use your data to build valuable insights about purchases and transactions.
• Reduce issues by using AI to predict problems for proactive action.

5. AI in finance: Budget smarter and boost security

AI-driven finance tools help startups budget, forecast, and detect fraud. Financial management is a critical AI use case for startups as it enhances budgeting accuracy and security.

Accurate budgeting and forecasting

AI tools can predict revenue, expenses, and cash flow, allowing you to create budgets that support growth while minimising risk.

• Create smarter budgets with reliable forecasts
• Drive growth decisions using valuable financial insights.

Real-time fraud detection

Data intelligence scans transactions to spot suspicious activity, adding an extra layer of security to protect your business.

• Receive instant alerts for irregular transactions
• Strengthen customer trust with reliable fraud detection

6. AI in Human Resources: Easy hiring and boosting employee satisfaction

For human resources, AI tools can improve recruitment and boost engagement, both of which contribute to a positive work environment. HR-related AI use cases for startups make the hiring process faster and more inclusive.

Simplifying hiring with AI

Make hiring easy by using AI agents to review resumes to find candidates with the right skills, speeding up hiring and reducing bias.

• Build inclusivity through skills-focused hiring.
• Shorten the hiring process with automation.

Increasing employee engagement

Boost your team morale with AI-backed analysis. AI analyses employee feedback to measure satisfaction, giving you insights to make a better team environment.

• Reduce turnover by increasing employee satisfaction.
• Enhance support through actionable insights for your team.

7. AI in security: Keep your data and customers safe

Startups often handle sensitive information, making cybersecurity essential. 72% of business leaders in Singapore cite privacy and security concerns as a barrier to purchasing AI, while 92% agree that trust is critical when partnering with an AI vendor. AI-driven cybersecurity is one of the most crucial AI use cases for startups, offering proactive threat detection. 

Spotting cyber threats

With AI monitoring network activity and user behavior, you can detect unusual patterns and prevent breaches before they happen, strengthening your business’s security.

• Monitor network security 24/7 with AI.
• Flag unusual patterns instantly with quick alerts.

Securing access

Stay secure with AI that controls who has access to data, making sure only authorised users can view sensitive information.

• Prevent breaches with controlled access to avoid unauthorised data use.
• Increase efficiency through automated access management to save time and reduce errors.

8. AI in product development: Stay ahead of market trends

AI assists in product development by spotting customer needs and market trends, helping you create products that customers want. Product development is an innovative AI use case for startups, enabling them to stay competitive.

Identify trends

Spot patterns with AI that can analyse data to reveal customer preferences and trends, helping you create products that meet demand.

• Match customer needs with trend insights to create desired products.
• Find opportunities by analysing trends to identify market gaps.

Optimise features based on feedback

AI can scan customer feedback to prioritise features, helping you develop products faster and more effectively.

• Streamline development with data-based choices to speed up timelines.
• Enhance satisfaction by prioritising popular features to keep customers happy.

State of the AI Connected Customer

Insights from 16,000+ consumers and business buyers worldwide on bridging the trust gap in an era of rising customer expectations and more powerful technologies.

Get the free report

9. AI in data analysis: Turn insights into action

Data analysis is a core AI use case for startups, allowing them to analyse large datasets, uncover patterns, and support data-driven decision-making.

Business intelligence and reporting

With an AI-powered solution like Data Cloud, you can analyse your data to generate meaningful reports, which helps to quickly identify opportunities and challenges — helping you stay competitive.

• Turn data into clear, strategic reports for actionable insights.
• Make informed decisions faster with AI.

Predictive analytics for various business functions

Predictive AI uses historical data to predict future outcomes, from customer behavior to sales performance. This helps startups make proactive, data-backed decisions.

• Anticipate trends and challenges to reduce uncertainty.
• Enhance forecasting accuracy for greater confidence.

Data visualisation and dashboards

AI-powered dashboards present data in a visually appealing and easy-to-understand format, making it easier to communicate findings and track key metrics in real-time.

• Track data in real time with visual dashboards that update instantly.
• Simplify data presentation for easier accessibility and understanding.

Final thoughts on AI use cases for startups

Your startup is ready to be powered by AI. By exploring AI use cases for startups and SMBs with tools like Salesforce’s Starter Suite, SMB Commerce storefront, and Agentforce, you’re setting up for a smarter, more connected future. This blog has shown how AI can make a difference in areas like sales, marketing, customer service, operations, security, finance, and HR.

With the right tools, you can get your customer data ready for AI with Starter Suite — try it out for free.

The post AI For Startups: 9 Use Cases For Growing Businesses appeared first on Salesforce.

As brands look for ways to get closer to consumers, more than half of marketers, 75% of marketers are already rolling up their sleeves and experimenting with or fully implementing AI. Our most recent State of Marketing survey found that top Marketing AI use cases are related to automation, highlighting the importance of scaling up speed and effectiveness. 

One click away from campaign success

Generative AI technology is transforming the future of marketing — and we’ve got you covered. Take a quick look on Trailhead, Salesforce's free online learning platform.

Kickstart your campaign

+100 points

Trail

AI Technology for Marketing

Let’s take a look at 9 ways using AI and agents as a digital assistant can increase the effectiveness and efficiency of your campaigns. It’s time to say goodbye to the redundant manual campaign tasks marketers wish they didn’t have to do – and let AI and agents help make the most of your time.

Table of contents

1. Make better decisions with automated data analysis and insights
2. Increase engagement and conversions with agent-driven audience segmentation
3. Anticipate your customer’s needs with predictive analytics
4. Save time with agent-driven end-to-end campaign assistance
5. Streamline workflows with campaign automation
6. Clearly show campaign success with performance tracking and reporting 
7. See what works best with A/B testing
8. Grow revenue with lead scoring and nurturing
9. Improve communication with internal collaboration tools

1. Make better decisions with automated data analysis and insights

AI can analyse large volumes of campaign data, including customer behaviour, campaign performance metrics, and market trends. It can identify patterns, extract insights, detect correlations, and provide actionable recommendations to improve campaign strategies and targeting.

You’ll get a deeper understanding of customers and campaign performance, enabling you to make informed decisions and find success faster.

Many brands are turning to AI agents, such as Agentforce, which uses secure data from your CRM to help you with tasks in a conversational way. For example, you can ask questions in the flow of work, just like you speak with a co-worker. Unlike chatbots and copilots, agents’ ability to plan and reason enables them to take action on the data, making recommendations and even doing things like build digital storefronts, draft custom code, and create data visualisations. And when needed, the agent can seamlessly hand off to human employees with a summary of the interaction, an overview of the customer’s details, and recommendations for what to do next. (Back to top)

2. Increase engagement and conversions with agent-driven audience segmentation 

After analysing the customer data, your AI agent can then segment audiences based on demographics, behaviour, preferences, purchase history, and other important attributes. AI eliminates the manual effort required for segmenting audiences and targets specific customers with more relevant offers. 

When you’re able to personalise messaging for different segments, you’ll see campaigns succeed more. (Back to top)

3. Anticipate your customer’s needs with predictive analytics 

AI predictive models use historical data to forecast customer behaviour, such as likelihood to convert, churn, or engage with specific campaign elements. 

This helps you stay one step ahead to proactively address customer needs and budget resources effectively. (Back to top)

4. Save time with agent-driven end-to-end campaign assistance

Creating a full end-to-end campaign with unique content frequently can be one of the most time-consuming tasks for many marketers — but an AI agent can help with using the right data foundation. Agentforce Campaigns powered by natural language prompts (NLP) and grounded by real-time data in Data Cloud can generate not only a campaign brief and target audience segment, but it can also save time creating content — such as ad copy, email subject lines, and social media posts — that resonates with your customers. 

You can provide the finishing touches to make sure the content is in your voice and tone. It can also improve your content by analysing performance data, identifying high-performing elements, and suggesting improvements. (Back to top)

5. Streamline workflows with campaign automation

AI and agents can automate various aspects of campaign execution, such as scheduling and deploying ads, sending targeted emails, or managing social media posts. This reduces manual effort and ensures that your campaign runs on time. 

What can you do with the time freed up, thanks to AI and agents? Focus on strategy and innovative ideas, helping you build lasting customer relationships. (Back to top)

6. Clearly show campaign success with performance tracking and reporting 

According to our State of Marketing report, high-performing marketers are able to analyse data in real time, giving them an advantage when it comes to responding to and optimising campaign performance.

Your AI agent can automate the tracking and reporting of campaign performance metrics — in ways that anyone can understand. AI can generate real-time dashboards and visually-pleasing customised reports, giving you and your stakeholders a clear view of campaign performance and key metrics, without the need to do it all by hand. 

This helps you make data-driven decisions, optimise campaigns on-the-go, and demonstrate the value of your efforts to stakeholders. (Back to top)

7. See what works best with A/B testing 

AI can perform A/B tests on campaign elements, such as ad variations, landing pages, or email designs. It analyses performance data, identifies winning variations, and helps you continuously refine your strategies. (Back to top)

Move faster with AI

Focus on innovation, not repetitive tasks. See how generative AI is transforming marketing.

Find out how

8. Grow revenue with lead scoring and nurturing 

With AI, you can automate lead scoring by analysing lead data, behaviour, and engagement history. It assigns scores to leads based on their likelihood to convert and deliver personalised content to move prospects through the sales funnel. 

With AI’s lead scoring, your team can focus on the most promising leads and nurture relationships at scale. (Back to top)

9. Improve communication with internal collaboration tools

AI shines as your digital assistant when handling internal collaboration needs. You can use this technology to automate messaging in your department, as well as project management, task assignment, and file sharing. Teams can even apply workflow automations that schedule meetings, send reminders, or organise files — taking care of the little details so you can focus on campaign success.

AI is transforming campaign management by allowing teams to automate manual tasks, freeing marketers to work on more big-picture ideas. With AI as your ally, you can streamline your campaigns, see better results, and start focusing on your next successes. (Back to top)

See what AI can do

Learn how AI can help you move more efficiently and create meaningful customer relationships.

Fuel your marketing

The post 9 Ways AI Can Save Marketers Time, Money — and Grief appeared first on Salesforce.

But in this fast-moving technical landscape, using personal data requires extra special handling. This is especially true when it comes to building and maintaining a trusted relationship with your customers, while remaining compliant with the growing number of regulations around the world.

Why data privacy compliance is non-negotiable

Law No. 27 of 2022 on Personal Data Protection (PDP Law) in Indonesia, Personal Data Protection Act (PDPA) in Singapore, Personal Data Protection Act (PDPA) in Malaysia, set standards for data protection.

Organisations face growing pressure to protect individual privacy rights while navigating an increasingly complex regulatory landscape. At the same time, consumers are more aware and concerned about how their personal information is collected, used, and safeguarded. By strengthening privacy and compliance frameworks, businesses can build trust with consumers and reduce legal risks and potential fines.

So, how do you keep customer data secure and protected while staying compliant with global regulations? Here are five steps to help you strengthen data privacy and compliance in your organisation.

Guard what matters most

From passwords to pixels, make sure that your organization is on top of the latest data privacy laws and regulations. See how on Trailhead, Salesforce’s free online learning platform.

Start learning

+400 points

Module

Data Privacy

1. Understand the data you have — and classify it

Categorising data by sensitivity — such as personal information, financial data, or health records — is crucial for effective data management and protection. It’s vital to understand what a piece of data is, how it can be used, and the protections around it. In turn, that allows you to implement targeted security measures and access controls that align with regulations. 

Proper data classification also helps pinpoint where sensitive information resides within an organisation. It sets you up to apply appropriate safeguards, such as encryption, pseudonymisation, or anonymisation.

This clear classification not only maintains compliance with data protection laws but also supports quick responses to data subject access requests, sticking to the principles of data minimisation and purpose limitation, all while creating a trusted relationship with your customers. 

Salesforce provides a free data classification tool that simplifies the process of classifying every standard and custom field, helping you identify your most sensitive data and incorporate it into your security and privacy policies.curity and privacy policies.

2. Audit and update your access controls

With your data classified, you can now assess who in your organisation should have access to what data. Audit your access controls and determine whether access rights are appropriate. 

Check if any accidental or intentional over-permissioning occurred, and review and update access permissions based on employee roles, data sensitivity, and regulatory requirements. Doing so can mitigate risks associated with data breaches and non-compliance. 

By proving the process of access control management, you’re prepared for a number of global regulatory inspections or audits while ensuring only those who absolutely need access to sensitive data can see it. You can manage access controls effectively by using tools that allow you to set permissions at various levels, ensuring that only authorised individuals can access sensitive data.

Salesforce allows you to manage access at a user, objective, and field level using the permissions and access settings. This approach helps maintain security and compliance across your organisation.

3. De-identify data in your testing environment

One way companies experience data breaches is by using real data in their testing environments. Everyone wants realistic data to test their application. But by anonymising or pseudonymising sensitive information, organisations can simulate real-world scenarios without compromising individuals’ privacy rights or experiencing a data breach. 

This practice ensures that any data classified as personally identifiable information (PII) in the first step (such as names, addresses, and social security numbers), is not exposed during software testing, reducing the risk of data breaches or unauthorised access. 

De-identification is key to data minimisation because it ensures you use only the necessary data for testing. This limits the risk of data exposure and keeps you in line with privacy laws. By using solid de-identification techniques and following ethical data practices, you can protect sensitive information and build stronger trust with your customers.

Solutions that protect sensitive data in secure testing environments, like Data Mask, are available to assist in de-identifying data. These tools can help you create policies to mask or replace sensitive information with non-identifiable data — using methods like random characters, similarly mapped words, pattern-based masking, or even deleted data. Pairing these tools with data classification (mentioned in the first step) ensures all your sensitive data is included.

Additionally, consider solutions that provide complete visibility into your testing environments and manage security. With tools like Security Center, you can centrally monitor, view, and manage your security health across multiple environments from a single platform, making it easier to maintain a strong compliance and security posture with actionable insights.

Data Foundations for the Age of AI

Get the free whitepaper

4. Set up monitoring and alerts on sensitive data 

With your data classified, access controls in place, and apps tested for privacy compliance, it’s time to set up monitoring, logging, and alerting systems to keep everything secure.

Tracking and logging user activities lets you keep an eye on access patterns, spot anomalies, and respond quickly to potential security issues. Proactive and real-time alerts can help you catch and block unwanted activity and can stop data leaks before they happen. 

By logging all of the actions in your system, you can research issues, learn from past behaviour, and improve monitoring management. Logging also sets you up to provide evidence of compliance during regulatory inspections or in response to data subject access requests.

Organisations can use toolsets to enhance compliance with data regulations and ensure data privacy. With tools like Event Monitoring, organisations can monitor security, track application performance, and glean product intelligence insights using event logs. 

It’s important to have solutions that proactively find security threats and respond effectively, respond to audits with ease by storing and querying event data using SOQL, and stay on top of compliance requirements.

5. Provide customer controls for consent and preference management 

Lastly, one of the most critical aspects of a privacy and compliance program is respecting your customers’ wishes for their data use. Complying with data subject requests, practicing data minimisation, and managing consent effectively are key to complying with global privacy laws.

Regulations emphasise individuals’ rights to access, delete, and revoke consent for their data. By quickly addressing these requests, organisations uphold privacy rights and avoid legal risks and fines associated with non-compliance. 

Implementing data minimisation ensures you collect and retain only the essential data, reducing the impact of potential breaches. And effective consent management means getting clear and informed consent before processing personal data, fostering transparency and trust. These practices will strengthen data protection and organisational credibility, showcasing commitment to ethical data handling practices in accordance with evolving privacy laws.

At the final step, consider solutions to help manage consent and data requests, allowing you to handle data privacy efficiently and maintain compliance. For instance, Privacy Centre is a suite of data management tools built to help you manage components of data privacy laws. It allows you to create, monitor, and track requests, automatically fulfilling data subject access and right-to-be-forgotten requests.

Customers can easily update their consent and preferences by hosting forms on your website or in Experience Cloud and updating their consent and preference data to your organisation, next-Gen Marketing Cloud, or Data Cloud. And you can de-identify, delete, or move personal and sensitive data.

With the right tools and practices in place, including de-identification, deletion, or relocation of personal data, you’ll maintain a classified, permission-minimised, and secure data environment, ready to tackle data privacy compliance with confidence.

Data governance for Agentforce

Unlock strategies for CIOs and CDOs to ensure data governance for Agentforce.

Get the free guide

The post Data Privacy Pitfalls? Not With These 5 Steps to Compliance Success appeared first on Salesforce.

Most importantly, generative AI technology can save time on tedious processes, so you can provide better care for your customers and focus on big-picture strategies. Let’s dig into how generative AI can help your business do more, and learn more about large language models.

What are large language models?

Generative AI is powered by large machine learning models that are pre-trained with large amounts of data that get smarter over time. As a result, they can produce new and custom content such as audio, code, images, text, simulations, and video, depending on the data they can access and the prompts used. 

To put things into everyday context, large language models provide answers depending on how a question is phrased. For example, “what are LLMs and how can they help my business?” versus “what are LLMs and what value can they bring to my business?” will yield different results. Although the questions are similar, responses can vary by context.

Because these models use natural language processing and machine learning capabilities, LLMs respond in a human-like, coherent, and relatable way. As a result, they excel in tasks such as text translation, summarisation, and conversations. 

With generative AI helping businesses perform these tasks, trust has to be at the core of your efforts. To make sure you’re using this technology responsibly, you can invest in a customer relationship management platform that has an AI-focused trust layer — which anonymises data to protect customers’ privacy. 

A trust layer built into a generative AI landscape can address data security, privacy, and compliance requirements. But to meet high standards, you must also follow guidelines for responsible innovation to ensure that you’re using customer data in a safe, accurate, and ethical way.

State of the AI Connected Customer

Discover how the growing use of AI, including generative AI and agents, is shaping customer sentiment, expectations, and behaviours.

Get the report

How do large language models work?

Advancements in computing infrastructure and AI continue to simplify how businesses integrate large language models into their AI landscape. While these models are trained on enormous amounts of public data, you can use prompt templates that require minimal coding to help LLMs deliver the right responses for your customers.

Furthermore, you can now create private LLMs trained on domain-specific datasets that reside in secure cloud environments. When a LLM is trained using industry data, such as for medical or pharmaceutical use, it provides responses that are relevant for that field. This way, the information the customer sees is accurate.   

Private LLMs reduce the risk of data exposure during training and before the models are deployed in production. You can improve prediction accuracy by training a model on noisy data, where random values are added in the dataset to mimic real world data before it’s cleaned. 

It’s also easier to maintain an individual’s data privacy using decentralised data sources that don’t have access to direct customer data. As data security and governance become a top priority, enterprise data platforms that feature a trust layer are becoming more important.

Businesses can also leverage how LLMs work with other kinds of AI. Imagine using traditional AI to predict what customers may plan to do next (based on data from past behaviour and trends), and then using a LLM to translate the prediction results into actions. 

For example, you can use generative AI to build personalised customer emails with offers, create marketing campaigns for a new product, summarise a service case, or write code to trigger actions such as customer recommendations. 

These large language models save time and money by streamlining manual processes, freeing up your employees for more enterprising work. 

Now that you’ve learned what generative AI can do, let’s see how you can use it to help your business. 

Learn about ethical AI

AI technology is only as good as the data it draws from. With this lesson on Trailhead, Salesforce’s free learning platform, discover how you can remove biases from your data and develop AI in your company responsibly.

Start here

+300 points

Module

Responsible Creation of AI

4 ways generative AI can help your business

The sky’s the limit when it comes to ways you can use generative AI for your business. 

LLMs are great at recognising patterns and connecting data on their own. Predictive and traditional AI, on the other hand, can still require lots of human interaction to query data, identify patterns, and test assumptions.

Feeding from customer data in real time, generative AI can instantly translate complex data sets into easy-to-understand insights. This helps you and your employees have a clearer view of your customers, so you can take action based on up-to-date information.

Now let’s dive into some use cases where large language models can help your business.

Using sentiment analysis to gain context into post-purchase actions

Sentiment analysis can help marketing, sales, and service specialists understand the context of customer data for post-purchase actions. For example, you can use LLMs to segment customers based on their data, such as using poor reviews posted on your brand’s website. These insights can help you act immediately on negative feedback. A great marketing strategy would be sending a personalised message offering the customer a special deal for a future purchase. This can help improve brand loyalty, customer trust, retention, and personalisation.

Generating email text for marketing campaigns

Text generation can help marketers reduce the time that they spend preparing campaigns. Generative AI can produce recommendations, launch events, special offers, and customer engagement opportunities for your social media platforms. Then, you can polish up the text to make sure it’s in your company’s voice and tone. For example, you can use the copy produced by generative AI to deliver personalised emails informing customers about a new product launch. This helps to improve personalisation, giving your customers a more consistent experience.

Surfacing related cases for service agents 

Case summarisation can help service agents to quickly learn about customers and their previous interactions with your business. Cases provide customer information such as feedback, purchase history, issues, and resolutions. Generative AI can help with recommending similar customer cases, so an agent can quickly provide a variety of solutions. This results in faster resolutions, time and cost savings, and happier customers. 

Automating basic code generation

Automation helps developers and integration specialists generate code for basic but fundamental tasks. For example, you can use code written by large language models to trigger specific marketing automation tasks, such as sending offers and generating customer message templates. This way, the overall language is consistent, personalised for the customer, and in your company’s voice. Automation can save time and improve productivity, allowing developers to focus on tasks that require more attention and customisation.

When used as part of a hybrid AI strategy, large language models can complement various predictive capabilities and drastically improve productivity. While generative AI can do so much, this technology still needs human guidance to be most effective for businesses. Generative AI can surface the insights you need to make decisions that can move your business forward. 

Think of it like a smart, automated assistant for your company, handling time-consuming tasks so your employees can work on complex problem-solving. When you blend the power of generative AI with the knowledge and expertise your company can provide, you’ll be able to do more for your customers.

Urvi Shah, Staff Technical Writer, contributed to this blog post.

The post What are Large Language Models (LLMs)? appeared first on Salesforce.

Welcome to the dawn of the Agentic AI Era. Starting now, almost any business—from individual contributors to executives—can orchestrate not just human workforces, but digital labour as well. We’ll see trust and accountability as the bedrock for an evolution unfolding in three stages: specialised agents mastering discrete tasks, multi-agent systems collaborating seamlessly, and enterprise-level orchestration rewriting how businesses operate. 

Salesforce’s AI Research’s role is to shape the future of enterprise AI. Here’s our vision for how agentic systems will advance, and what will be needed from humans to help them along the way.

The Evolution of AI Agents: From Rules to Reasoning

The progression of AI agents mirrors the development of machine learning itself. Traditional rule-based systems like Robotic Process Automation (RPAs) were capable of executing precise sequences but stumbled when faced with variations. These early implementations required substantial technical overhead and consulting services, creating a high barrier to entry for many organisations.

The last few decades have witnessed incremental and breakthrough advances that have transformed how machines process information—evolving from rigid automation to more flexible, adaptive, and far more efficient learning systems. Agents built with modern platforms like Agentforce can understand context, adapt to new situations, and handle broad task spectrums. But what’s even more exciting is where we’re headed: self-adaptive agents enabled by multi-agent reasoning—agents that can learn from their environment, improve through experience, and collaborate both with humans and agents from our enterprise customers, partners, vendors, and even the personalised AI assistants of consumers, which are becoming a bigger part of their lives every day.  We are only at the beginning of a three-stage future for Enterprise AI agents. 

Three Stages of Enterprise AI Agents

Just as music evolved from single-note melodies to complex symphonies, AI agents are progressing from solo performers to orchestrated ensembles. Each stage builds upon the last, creating richer, more nuanced interactions in the enterprise environment.

Stage 1: “Monophonic” AI – The Specialised Contributor

In the first stage of agentic evolution, specialised agents excel at defined tasks within particular industries, bringing unprecedented efficiency and accuracy to routine but crucial business operations. They represent the foundation of enterprise AI adoption, handling discrete tasks with a level of consistency and speed that transforms departmental workflows. They also are masterful at providing the benefits of AI’s advancements to date, like predictive next best actions and product recommendations, highly personalised to each customers’ preferences and behaviours. And generative guidance, marketing language and correspondence of the highest caliber, for customers, service and sales reps—humans and bots alike. 

In commerce, for example, they revolutionise inventory and account management. Indeed, agents don’t just handle basic inventory checks; they proactively monitor stock levels across multiple locations, predict seasonal demands, and generate real-time account summaries that flag unusual patterns or opportunities. Tasks that once required hours of human analysis can now be completed in seconds, with greater accuracy and depth, yielding optimised, personalised, and almost “magical” experiences for the retail customer. 

Service operations see similar transformations. Beyond basic billing summarisation, these agents analyse customer interaction patterns, automatically categorise, and prioritise service requests, and generate predictive insights about customer needs. They spot trends in customer behaviour that might indicate satisfaction issues or expansion opportunities, providing service teams with actionable intelligence rather than raw data. The result is customer service that feels effortless, ambient and almost invisible to the end customer – their issue is now often resolved before they even knew there was one. 

In financial services, agents redefine customer service efficiency. When processing dispute acknowledgments, they analyse transaction histories, identify patterns of potentially fraudulent activity, and automatically trigger relevant security protocols. For financial planning, they generate comprehensive analyses by correlating market data, individual client histories, and broad economic indicators. When used correctly, these agents will afford businesses unprecedented back-office efficiency, and will inform next-generation retail banking, investment guidance and wealth management experiences for consumers.

Stage 2: “Polyphonic” AI – The Seamless Collaborators

This stage introduces orchestrated collaboration between specialised agents within the same company, collaborating together toward a common business goal. In this case, an “orchestrator agent” coordinates multiple specialists working in concert, similar to how a restaurant’s general manager orchestrates talented hosts, servers, managers, chefs, prep cooks, and expediters to work together to earn that coveted Michelin star.

What does polyphonic AI look like for a complex business operation? Consider a customer service scenario where multiple agents work invisibly together to support a loyal retail customer’s request ticket to exchange sizes of an off-season SKU. 

• A front-line service agent processes the initial customer enquiry
• An inventory specialist checks product availability across locations
• A logistics agent calculates shipping options and timelines
• A billing expert reviews account history and payment options, and most importantly:

The orchestrator agent coordinates all these inputs into a coherent, effective, on-brand and contextually relevant response for the human at the helm to review, refine, and share with the customer.

When implemented well, this multi-agent approach, with an “orchestrator agent” serving its “orchestrator human” delivers powerful AI-driven advantages: The system achieves enhanced reliability by leveraging specialised, trusted agents focused on specific domains, while reducing hallucinations since each agent operates within a narrower scope. This distributed approach also strengthens security by isolating sensitive data handling to specific agents. Perhaps most importantly, the ecosystem offers seamless scalability—organisations can continuously add new specialised agents to expand capabilities as needs evolve.

Stage 3: “Ensemble” AI – The Enterprise Orchestrators

The final stage—the ideal stage—adds sophisticated agent-to-agent (A2A) interactions across organisational boundaries, creating entirely new patterns of business relationships. Beyond traditional B2B and B2C models, we see the emergence of B2A (business-to-agent) and even B2A2C interactions where AI agents serve as intermediaries for work and transactions.

Consider a simple car rental scenario: A customer’s personal AI agent negotiates with a rental company’s business AI agents. The customer’s agent optimises for the best price and value, while the rental company’s agent aims to maximise revenue through add-on services. But the business agent must balance aggressive sales tactics against the risk of losing the deal to competitors. These interactions can be governed by sophisticated “game theory” principles, requiring advanced negotiation skills and protocols, risk management under uncertainty, verification mechanisms to ensure trust along the way, not to mention the ability to deftly resolve conflict.  

Now, imagine this scaling to ever-more complex enterprise processes we see across industries: from supply chain optimisation to customer experience orchestration. Whether you’re a consumer or enterprise employee, ensemble AI will mean that you’ll have an assistant to perform complex orchestration and meaningful collaboration per your personalised needs and wishes. And in order to achieve this, we as humans have some work ahead of us. 

Non-Negotiable Imperatives: Trust and Accountability

As we deploy increasingly sophisticated agent systems, two fundamental principles must guide every decision: trust and accountability. 

Building Trust

Trust in the era of agentic AI extends far beyond technical safeguards against toxicity, bias, and hallucinations. Recent Salesforce research shows 61% of customers believe AI developments make trustworthiness more critical than ever‌—and they’re right. We’re entering territory that demands deep organisational confidence in the symbiotic relationship between humans and AI.

This confidence builds on four essential foundations.

First is the bedrock of accuracy and boundaries‌ — ‌AI agents must operate within well-defined parameters while maintaining precision. Beyond preventing errors, these guardrails will create predictable, trusted partnerships that amplify collective intelligence.

Just as crucial is an agent’s self-awareness. Like any valued colleague, AI agents must acknowledge their limitations and know when to engage human expertise. This requires sophisticated handoff protocols that ensure seamless collaboration between artificial and human intelligence. For example, our AI Research team explores training methods to teach AI agents to flag areas of uncertainty and seek assistance when confronted with unrecognised challenges. Trained correctly, AI will know when not to attempt a guess but rather to come to a human and ask for help. 

For multi-agent systems, we will also need engagement protocols that are globally accepted and adopted. Think of it like this: Banks have global protocols, or rules to systemise the transfer of funds between individuals, businesses and countries. Traffic has protocols to ensure adherence to rules, governed by our universal traffic light colour system. The Internet has “IP” – our global Internet Protocol that allows for routing and addressing of packets of data to travel across networks and arrive at the correct destination.

So too will agents of the future need these protocols that are agreed upon and implemented universally, so that orchestrator agents can communicate, negotiate and collaborate with other business’s agents safely, ethically and for mutual benefits of both parties. This “ensemble” level of engagement will need to be fast, efficient and fair. Without such protocols in place, we’re at risk of agent-to-agent “spam” at best, and fraud and other dangers at worst. 

Finally, as our AI agent workforce grows, so must our security measures. As with any technology, humans with malicious intentions can also wield AI, designing and training “AI Worms” for the purposes of data breaches or to attempt to hijack other AI agents to disclose private customer data. Enhanced protection, privacy controls, and continuous monitoring mustn’t be seen as mere technical requirements—they’re essential to maintaining the trust that transforms AI from a tool we use into a partner our businesses will grow with.

State of the AI Connected Customer, 7th edition

Discover how the growing use of AI, including generative AI and agents, is shaping customer sentiment, expectations, and behaviours.

Get the report

Ensuring Accountability

As organisations deploy AI agents that make thousands of decisions per second, we must establish clear frameworks for responsibility and oversight to ensure we have a plan for if and when things go wrong. This requires a comprehensive approach. Below is a starting point for C-Suite teams overseeing an agent implementation effort.

• Clear chains of responsibility for agent decisions. When an AI agent makes a consequential decision, there should be no ambiguity about who’s accountable. This may even mean establishing new roles like “AI Operations Officers” who have both the authority to oversee agent deployments and the responsibility when issues arise.
• Robust systems for detecting and correcting incomplete information, biases, hallucinations, or toxic outputs—before they impact your business. This goes beyond basic safety checks to include continuous monitoring of agent decisions, real-time intervention capabilities, and systematic audit trails. Just one example of this is our research team’s recent advancements in retrieval-augmented generation (RAG), dramatically improving how our AI systems access and verify information. These innovations enable rapid evaluation and course-correction—ensuring that AI systems deliver accurate, reliable results that humans and businesses can trust.
• Defined processes for human oversight and intervention that balance autonomy with control. We need to move past the simple notion of “human in the loop” to develop sophisticated frameworks for when and how humans should intervene in agent decisions. This means creating guidelines and org-wide standard ways of communicating with agents as well as clear escalation pathways that maximise agent autonomy for routine tasks while keeping human judgment central to high-stakes decisions.
• Structured approaches for making things right when mistakes occur. This includes not just technical rollback procedures, but also clear protocols for customer communication, remediation, and systematic improvements to prevent similar issues.
• New legal and compliance frameworks that explicitly address AI agent accountability. The current regulatory landscape wasn’t designed for autonomous AI agents making business decisions. We need to work proactively with regulators to develop appropriate governance structures. proactively with regulators to develop appropriate governance structures.

Deploy AI agents with confidence

Fast-track your work with AI and human collaboration to drive a long-term success.

Learn about Agentforce

Looking Ahead: The Scientific Method Meets Enterprise Innovation

The path to deploying truly interactive AI systems demands executive foresight: we must apply the same stringent scientific standards that produced these advances to their real-world implementation. Success won’t be just decided by the number of AI agents deployed or implementation speed, but by how thoughtfully enterprise leaders and technologists orchestrate their integration with existing workforce protocols, processes, and preferences.

As we advance our understanding of agent collaboration, shared learning, and human-AI interaction, we’re discovering principles backed by reproducible research and empirical evidence. Drawing on Salesforce’s decades of enterprise CRM success and expertise in business logic and optimisation, we’ve infused Agentforce with deployment strategies that ensure our systems are not just powerful, but trustworthy and accountable in meeting the needs of our customers’ business—and the humans that run them.

The future isn’t about humans versus AI – it’s about humans with AI working in concert, each using their unique strengths. Agents will become—and with the launch of Agentforce, indeed already are— a true workforce multiplier, enabling teams to tackle previously impossible tasks. The time to begin this transformation is now, and the scientific method will light our way forward: through careful hypothesis testing, meticulous measurement, and continuous refinement based on evidence. Just as every breakthrough experiment begins with a hypothesis, every successful AI transformation begins with a vision—and ends with validated truth.

Resources

• Salesforce AI Research website
• Follow us on X: @SFResearch

The post The Agentic AI Era: After the Dawn, Here’s What to Expect appeared first on Salesforce.

Artificial intelligence (AI) is transforming the job market at lightning speed. AI has the potential to create new job opportunities in Southeast Asia. And these new jobs aren’t just plentiful. They’re rewarding. In Vietnam, for instance, AI engineers are among the top earners in Vietnam’s IT industry with salaries of US$1,110-2,060 a month, excluding bonuses.

AI won’t replace you, but someone skilled in working with AI could gain a competitive edge. That’s why developing the right skills to collaborate with AI is more critical than ever. To help you stay ahead of the curve, we’ve curated this list of the top 10 human-AI collaboration skills you need to thrive in the workplace.

These skills will help you develop effective human-AI partnerships. By understanding AI, its role in the workforce, and the skills needed for success, you can prepare for the days and years ahead.

The benefits of human-AI collaboration in the workplace

But before we explore the skills of the future, it’s important to understand why the ability to work effectively with AI matters so much.

AI can make work faster and more efficient in areas like sales, service, marketing, inventory management, and factory operations.

State of the AI Connected Customer, 7th edition

Discover how the growing use of AI, including generative AI and agents, is shaping customer sentiment, expectations, and behaviours.

Get the report

The future of work is changing, but it doesn’t mean humans are fully replaceable. George Hanson, the Chief Digital Officer at Mattress Firm, put it this way during an interview on Experts of Experience, “The value I see in AI is as an aid to humans, as opposed to replacement of humans.”

AI is revolutionising the way we work by enhancing efficiency, increasing productivity, and uncovering insights that might otherwise go unnoticed. However, behind every AI success story lies human effort and ingenuity. While AI tools can augment ‌ — ‌ and sometimes even replace ‌ — ‌ certain tasks, the real magic happens with human guidance. It’s people who train these systems, collaborate with them, interpret their outputs, and ultimately make the final decisions. At its core, workplaces still rely on human judgement, creativity, and the unique perspectives only we can bring.

What this means for you is there’s a massive opportunity to stand out, enhance your skills, and drive exceptional results. By fostering human-AI collaboration, you can combine your unique strengths with the capabilities of the technology to achieve remarkable outcomes. 

Top 10 collaborative skills for human-AI teams 

Below are the most important technical, analytical, and soft skills needed for the jobs of the future. Keep ahead in the changing job landscape by adopting these skills. You’ll also be able to boost your team’s productivity and elevate your career to new heights.

Skill 1: Understanding Generative AI

Develop a foundational grasp of generative AI—how it works, its capabilities, and its limitations. You don’t need to be a programmer, but knowing the basics is essential for effective collaboration. Focus on these key areas:

• Large Language Models (LLMs): These models, like ChatGPT, excel at creating text, summarising content, drafting emails, and answering questions.
• Machine learning systems: Perfect for analysing large datasets, making predictions, or finding patterns, like forecasting customer demand or personalising recommendations.
• Limitations and Ethical Use: Understand the boundaries of generative AI, including biases, data privacy concerns, and when human oversight is critical.

Skill 2: Prompt engineering

Think of prompt engineering as having a conversation with the AI‌. ‌It’s all about asking the right questions in the right way. AI systems work best when given clear instructions. Plus, you’ll get trusted results with accurate, relevant prompts grounded in your own data. Learn how to define exactly what you need, provide context, set boundaries, or use examples. Only then can you ensure the AI delivers useful and accurate outputs for your tasks. 

Skill 3: Familiarity with AI tools and platforms

Competence with popular AI platforms and tools is important. Staying up-to-date with new AI technologies helps you adapt quickly and improve your use of AI systems. Keep a close eye on AI developments in your industry. Subscribe to relevant publications, attend conferences, and follow thought leaders in the AI space.

Discover Agentforce

Agentforce provides always-on support to employees or customers. Learn how Agentforce can help your company today.

See the benefits

Skill 4: Judging the credibility of an answer

AI-generated insights can’t always be trusted at face value. It can sometimes generate information that sounds convincing, but isn’t entirely accurate. That’s why important to learn how to assess their relevance and accuracy and identify potential biases. Think of yourself as a fact-checker for your AI assistant. It’s crucial to evaluate its responses critically, verify facts, and stay aware of possible inaccuracies. By taking an active role in validating AI-generated insights, you can make more confident, informed decisions.

Skill 5: Knowing what problem to solve 

AI isn’t the answer to every challenge‌ — ‌you must know when to use it. AI shines with repetitive, time-consuming, or data-heavy tasks, helping you work faster and smarter. But for creative thinking or complex decisions, people are still the experts. When you understand what AI is good at (and what it’s not), you can let it take care of the busy work, giving your team more time to focus on higher-level work.

Skill 6: Data literacy

Being data-literate simply means understanding the basics of how data is structured and how it’s used. AI systems rely heavily on data for training, analysis, and decision-making, so understanding how data works is crucial to use and interpret AI outputs. You should be familiar with:

• Different types of data: Learn the difference between structured and unstructured data. Think neatly organised rows in a spreadsheet versus a collection of customer reviews or social media posts.
• Data-gathering methods: From surveys to web scraping, knowing how data is collected helps you gauge its reliability and relevance.
• Data interpretation techniques: Spotting trends, comparing metrics, and asking questions like, “What story is this data trying to tell?”

Data Literacy Basics

Understand the importance of data literacy and how to work effectively with data.

Visit Trail

+300 points

Module

Data Literacy Basics

Skill 7: Adaptation and flexibility

AI is always evolving, so you need to be flexible and continuously upskill yourself to keep pace with technological advancements. Developing a growth mindset and embracing lifelong learning is key. 

Invest in your own learning by following thought leaders, keeping up on trends, and exploring training programmes that help you develop the skills needed to work effectively with AI. Platforms like Trailhead offer a fun and engaging way to upskill, with an extensive library of AI-powered learning tools to get you started.

Skill 8: AI translation

Insights are useless if they can’t be shared. A critical skill is distilling AI-driven information into clear, actionable takeaways that customers and partners can easily understand. Mastering the art of translating AI outputs will significantly boost your value and ability to help your organisation embrace AI.

Skill 9: Curiosity and experimentation

While AI excels at automation and supporting roles, uniquely human skills like creativity remain indispensable. Cultivate a mindset of curiosity‌ — ‌questioning assumptions, exploring why things are done a certain way, considering whether systems can be improved, and thinking creatively about implementing new approaches. With AI handling many of your routine tasks, use the freed-up time to experiment, innovate, and try new ideas. 

Skill 10: Ethical judgment 

AI systems are built by humans and are influenced by the biases and moral perspectives of their creators. This means you have to rely on your own ethical judgement to determine whether something is appropriate or responsible to move forward with. As AI becomes widespread, understanding its ethical implications is essential. You’ll need to make real-time decisions about when, where, and how to use it responsibly. 

Deploy AI agents with confidence

Fast-track your work with AI and human collaboration to drive a long-term success.

Learn about Agentforce

A final thought on AI education

The future of work isn’t about humans versus machines. It’s about reskilling and learning how to work with AI to do incredible things together. By building these 10 essential skills, you’re not just preparing for what’s ahead, you’re setting yourself up to thrive in a world full of new possibilities.

Think of AI as your sidekick, not your replacement. Human-AI collaboration is here to make your job easier, free up your time, and help you focus on what you do best. With a curious mindset, a willingness to learn, and a focus on using AI responsibly, you can unlock opportunities that take your career to the next level.

The post The Future of Work: AI Adoption and 10 Human-AI Collaboration Skills appeared first on Salesforce.

Artificial intelligence (AI) in HR is more than just a trend. By integrating AI into their work systems, companies can automate routine tasks, deliver better employee experiences, and gain viable insights that can help improve HR operations. Here are three ways AI in HR is shaping the workplace, and how you can get started.

What is AI in HR?

AI in HR is the integration of AI technologies into HR processes. This includes using machine learning algorithms, AI agents, natural language processing (NLP), and data analytics to automate repetitive tasks, such as answering common employee questions like:

• What are my options for benefits? 
• What items are eligible for reimbursement? 
• What’s our expense policy? 

AI can also provide valuable insights into employee behavior and trends, aiding in strategic decision making. By using AI, HR professionals can focus on high-value activities like employee development and engagement, ultimately creating a more streamlined and effective HR function.

What are the benefits of AI in HR?

With AI in HR, teams become more efficient and productive. An AI in HR platform like Salesforce Employee Service helps you achieve the following.

Connect employees to the right answers fast 

This is one of the most significant benefits of AI. For example, in an employee portal, workers can use a generative AI-powered search feature to instantly access a wealth of information without needing to read through each article. Whether they need to understand company policies, check their benefits, or learn about upcoming training opportunities, employees will quickly get the information they need to perform their jobs effectively.

Set up HR teams to support employees faster

AI also helps HR teams provide faster and more efficient support to employees. With an AI-driven HR service console, for example, HR professionals can automate routine tasks such as answering frequently asked questions, processing requests, and managing employee records. This frees up HR teams to focus on more strategic and high-value activities, like employee development and engagement. By reducing the time spent on administrative tasks, HR teams can offer more personalised and timely support to employees to improve overall job satisfaction and productivity.

Boost employee and HR team efficiency with AI agents

AI agents are another powerful tool in the AI in the HR arsenal. They can handle a wide range of employee inquiries and requests, from scheduling meetings to providing guidance on complex HR policies. And with prebuilt integrations into systems like Workday, AI agents like Agentforce can understand and respond to even complex employee queries in a humanlike manner, providing accurate and helpful information around the clock. This not only boosts employee efficiency by giving them instant access to the support they need, but reduces the workload for HR teams and lets them focus on more critical tasks.

How to get started with AI in HR

Integrating AI into your HR processes can seem daunting, but by following a few easy steps, you’ll be on your way. Here’s how to get started:

1. Locate your HR content

Gather all your HR-related content, including policies, benefit plan summaries, standard operating procedures, employee handbooks, and any other relevant documents. Don’t worry if this content isn’t already in Salesforce — you can import content from various repositories like SharePoint into Salesforce Knowledge or Data Cloud. Centralising your HR content is crucial for creating a comprehensive and accessible knowledge base.

2. Start building your service catalogue

Next, think about the services and requests you want to offer employees. Identify the most common requests and determine if they can be turned into catalogue items that employees can request via the portal. For example, requests for leave, benefits changes, or a corporate credit card can be automated: After an employee provides the required information, a case would be created and routed to the appropriate fulfillment team.

3. Decide on the right channels

Employee Service can be delivered through multiple digital channels, including a portal, email, AI agents, web chat, WhatsApp, and more. Consider which channels are right for your organisation. Different channels cater to different employee preferences and can enhance the overall employee experience.

4. Start building your business case

To get your leadership on board to invest in AI for HR, you’ll need a strong business case. Begin by estimating how much time you could save your HR specialists and business partners if common inquiries and tasks were handled by AI. Consider what strategic projects these specialists could focus on with that extra time. Finally, calculate how reduced case-handling time and case age could decrease the cost to serve, making your operations more efficient and cost-effective.

Succeeding with AI in HR

AI in HR is transforming how we work. It simplifies finding answers and completing tasks, boosts HR team productivity by automating routine tasks, and uses AI agents to handle mundane tasks efficiently.

AI Strategy Guide

Whether you’re just starting out with AI or already innovating around the technology, this guide will help you strategise effectively, embrace new possibilities, and answer important questions about the benefits of AI.

Get started now

The post 3 Ways AI in HR Is Shaping the Employee Experience — And How To Get Started appeared first on Salesforce.

A game-changer for businesses, AI agents allow them to boost productivity, deliver personalised customer experiences , and drive topline growth. But what’s next? 

What else can businesses in ASEAN expect, and how can they leverage AI agents to grow their business and develop their workforce further? 

We got Sujith Abraham, SVP and General Manager for Salesforce ASEAN, and Gavin Barfield, Vice President & Chief Technology Officer, Solutions, Salesforce ASEAN, to share their top 10 predictions that will shape the course of ASEAN’s dynamic business landscape in 2025. 

Sujith leads the Salesforce ASEAN business and brings more than 20 years of leadership experience in enterprise technology to his role. He’s passionate about ensuring the success of our ASEAN customers, partners, employees and community, and transforming good ideas and people into impactful businesses built on strong culture and high levels of engagement. Gavin also has over 20 years of experience, with a solid technology background that includes IT infrastructure, enterprise architecture, cybersecurity, and program management across a variety of industries. 

Given their expertise and authority in the field, they foresee ASEAN businesses transitioning from AI experimentation to full-scale implementation in 2025, as businesses work toward a future where humans and agents drive customer success together with AI, data and action. Here are the key trends identified.

1. AI progressing beyond the experimentation stage, driven by autonomous agents

When Generative AI launched in 2022, there was huge excitement around the technology’s potential to revolutionise operations, boost productivity, and enhance customer experiences. Many businesses invested significantly in developing a generative AI strategy.

Despite initial excitement, few organisations have moved beyond Proofs of Concept (POCs) and limited trials to full-scale implementation. In some cases, Generative AI has failed to deliver accurate and useful outputs due to incomplete data. In others, solutions are disconnected from workflows, making them clunky and inefficient. Many applications of Generative AI, such as copilots and chatbots, were created as “solutions looking for problems”, focusing on experimentation rather than solving actual business issues.

Unlike chatbots and copilots, AI agents can autonomously navigate tasks and make real-time decisions directly in the flow of work — moving from mere assistance to taking action based on live data and context, marking a major step forward in enterprise AI.

In 2025, purpose-driven AI agents designed to address specific workflow needs and provide measurable benefits will help organisations move beyond experimentation to achieve tangible outcomes. For this to happen, generative AI needs to be grounded in the right data and delivered in the flow of work to offer meaningful impact.

2. Autonomous agents providing opportunity for topline growth

In the past two years, businesses have focused on cost-cutting measures in response to global economic uncertainties and slowing growth. The availability of autonomous agents today has now surfaced more opportunities for businesses to drive topline growth. 

How, you ask? These agents will deliver on the promise of AI, succeeding where solutions such as copilots fall short. One limitation of earlier innovations was their siloed focus on unstructured data. For example, copilots could only act based on data within emails or presentations. This misses critical transactional context, such as customer purchase history or product details. 

Copilots could only see part of the customer story, causing them to fall short on providing actionable insights. Businesses, in turn, missed out on opportunities to foster deeper and more compelling customer relationships that generated new revenue streams.

Autonomous agents can significantly impact a company’s growth trajectory. Take a bank that works with thousands of business clients as an example. An initial analysis of spend may lead the bank to think that most of their customers are SMEs with small spends. But a deeper look reveals that these businesses are spreading their spend across banks. 

It’s extremely difficult to turn the workforce around to deepen engagements with all customers. Imagine if they implemented autonomous agents to maintain consistent customer engagement without constant human oversight. And agents operate 24/7 – think of how much coverage across customers is now possible. This enables the bank to increase its revenue base, which might otherwise be lost to competitors.

AI agents also allow sales teams to automatically pre-qualify leads before handing them over to human agents. This way, human agents don’t waste time on unresponsive prospects, basic inquiries, or low-engagement leads, which can significantly improve productivity and the bottom line.

Deliver seamless experiences
with Agentforce

Build autonomous agents that work side-by-side with your employees to extend your workforce and serve your customers 24/7.

See how it works

3. Out-of-the-box AI/agentic solutions and unified data will underpin AI success

In the race to operationalise AI, the winners will be those who forgo DIY solutions in favour of out-of-the-box solutions that offer superior speed, deployment, and accuracy. Businesses that adopt out-of-the-box solutions can focus on AI deployment and achieve immediate impact and value. In contrast, those who attempt to “DIY” their AI often face setbacks in the form of hidden costs and a slow realisation of AI capabilities. 

Having the right data foundation is also key to maximising ROI from AI investments. Businesses need to consolidate structured data, such as customer transaction records, and unstructured data, such as customer emails, product information, and corporate policies, to build a unified view of their customers. 

Without it, AI cannot deliver accurate, contextualised, and trusted outputs. Zero-copy capabilities are needed to ensure companies maximise their existing assets while minimising data preparation costs.

4. An organic AI ecosystem emerging within the region

AI is ushering in one of the biggest technological shifts of our generation, creating new services, roles, and industries. Just as the invention of smartphones and mobile applications created a thriving ecosystem of app developers, the growth of AI platforms is fostering a new generation of AI developers. 

This drives innovation in ASEAN and opens the pathway for local talent to develop AI tools tailored to meet the region’s unique needs — whether it is Small Language Models (SLMs) that support native languages like Singlish or Taglish, or advanced models that tackle specific challenges like anti-money laundering. 

With a combined population of over 650 million (including individual markets with over 100 million people such as Indonesia and the Philippines), and a combined GDP comparable to major economies, there’s a massive opportunity for AI developers in ASEAN. 

The growth of the AI industry in the region will not only attract established global tech giants to set up operations and company headquarters locally, but also catalyse the birth of home-grown startups. 

With that, we’ll see a migration of strategic roles typically available in the West to this part of the world, creating new opportunities for the future workforce.

5. AI agents disrupting traditional service models in ASEAN with scalable capacity, intelligence, and personalised experiences

In ASEAN, businesses often hire additional service staff as a quick fix for improving customer experience, especially with lower labour costs in the region. However, increasing headcount alone doesn’t necessarily improve problem resolution or overall customer satisfaction.

AI agents provide a fundamentally different approach by autonomously handling requests and enhancing customer interactions in ways that go beyond scaling capacity. This isn’t about efficiency alone but delivering high-quality customer service. AI agents leverage real-time data to provide context-aware support, making decisions and taking action as customer needs arise.

With AI agents embedded directly into workflows, businesses can reimagine customer service, delivering faster and more accurate responses without increasing complexity or requiring extensive training. 

6. AI agents redefining jobs, empowering employees to focus on strategic work

AI agents are transforming the workforce by automating repetitive and time-consuming tasks, freeing employees to focus on higher-value work that drives innovation and growth. 

This presents an opportunity for the workforce to transform their skill sets and take on more strategic roles. As AI agents become increasingly integrated into the workforce, employees will need to develop new skills to manage and optimise them. They’ll also have to leverage their industry knowledge to train these agents so that they can deliver the desired business outcomes. 

This transformation mirrors the 1980s shift in banking, when staff moved from routine tasks like producing bank statements to customer service and financial advisory roles as automation took over.

Fast-forward to today — a telco that relies on AI agents to handle routine customer inquiries at scale will be able to empower its employees to focus on strategic tasks like optimising AI deployment and enhancing customer experiences. 

This not only creates new career opportunities for the telco’s staff, but also allows them to save on operational expenses and infrastructure costs. With virtual AI agents running routine operations, the telco can easily scale operations, without necessarily building new offices or stores for its workforce to service the expanded client pool. 

By blending human expertise with AI, companies can create a more agile workforce focused on driving growth and preparing employees for roles that require creativity, problem-solving, and strategic thinking.

7. New AI skill sets required for building and testing agents, including defining guardrails

As AI agents become central to business operations, there’s a greater need for professionals with specialised skills to guide these systems effectively. These skills will include being able to define agent instructions, craft prompts, and set guardrails.

Writing prompts may seem straightforward because they are written in natural language. However, crafting and refining these instructions and establishing clear guardrails to ensure an AI model performs as intended requires expertise.

While prompt engineering for LLMs is common, writing instructions and setting guardrails for reasoning engines will become critical skills. As more organisations integrate AI agents into their workflows, the demand for professionals with the skills to build and test agents in real-world scenarios will increase.

8. New types of AI models will push the boundaries of what AI can deliver

In 2025, we’ll see new, highly specialised AI models that go beyond text generation to drive complex, autonomous actions. Salesforce’s xLAM (Large Action Model) is at the forefront of this evolution. Unlike traditional Large Language Models (LLMs), which excel at generating responses, xLAM models are designed for action and decision-making, allowing AI to autonomously execute tasks and manage workflows without requiring explicit instructions.

These Large Action Models add a new dimension to CRM, enabling AI agents to handle tasks like function-calling, reasoning, and planning, adapting actions to fit real-world business contexts. By managing entire workflows proactively, like an autonomous sous chef that prepares each step, xLAM models can streamline operations and enhance decision accuracy across various environments.

As businesses adopt these models, xLAM can operate across multi-agent systems, coordinating actions between specialised AI agents to tackle increasingly complex, customer-focused processes. This innovation will make AI a powerful partner in business, delivering efficiency, context-aware responses, and automated actions that drive customer success with accuracy and reliability.

We’ll also see a proliferation of Small Language Models designed for particular industries or purposes.  These models are trained on smaller but more reliable datasets and are effective at performing certain tasks.  They’re cheaper to run, train and often more accurate than ‌Large Language equivalents.

9. Agents building agents, agents talking to agents becoming commonplace

Just as organisations have employees specialised in specific functions, AI agents will soon be assigned unique roles within a network. These agents will work alongside human employees, communicate with other agents, and create new agents as business needs evolve. Each agent will have a defined function, allowing the network to handle a wide range of tasks efficiently.

In this agent network, meta-agents will be crucial, coordinating actions across other agents to keep workflows seamless. For example, a concierge agent might interact with users, guiding them on tasks it can help with and providing updates on task progress. 

An orchestration agent would assess user needs and route requests to the proper agent, ensuring tasks are managed effectively. This setup enables collaboration on platforms like Slack, where human employees and AI agents can interact as a unified team, improving responsiveness and coordination.

10. Robotics driving the next wave of AI innovation

Robotics, the fourth wave of AI will emerge, transforming how businesses and customers connect. Beyond agents, robotics will see interactions evolve from text and voice systems to immersive experiences with physical robots and virtual avatars with lifelike, dynamic, and highly interactive engagements.

Picture virtual avatars powered by AI agents, with heads that move, lips that smile, and expressions that react naturally during interactions. This evolution will create more personal and engaging experiences in physical settings like a concierge in shopping malls, where customers can hold a real-time conversation with a lifelike avatar rather than typing queries into a screen.

At the same time, physical robotics will bring AI agents into the tangible world, unlocking new opportunities in environments such as F&B. Imagine a robotic barista powered by an AI agent that can offer personalised drink recommendations based on a customer’s past orders‌ — ‌including details like sugar preferences‌ — ‌and prepare the drink instantly.

Robotics will empower businesses to deliver natural, lifelike, and highly personalised interactions, redefining customer experiences that are powered by AI agents

Take your business to the next level with Agentforce

Now that you know what’s in store for ASEAN businesses in 2025, why not welcome Agentforce into your business so it can unlock unlimited potential with your team? 

Agentforce is the agentic layer of the Salesforce platform for deploying autonomous AI agents across any business function — enabling you to scale your workforce. Build and customise autonomous AI agents to support your employees and customers 24/7, and access a library of ready-to-use skills for any use case across sales, service, marketing, commerce, and more.

Our handy ROI calculator can help you measure Agentforce’s value for your business, and you can take a closer look at how agent building works here.

AI Strategy Guide

Whether you’re just starting out with AI or already innovating around the technology, this guide will help you strategise effectively, embrace new possibilities, and answer important questions about the benefits of AI.

Get started now

Read more:

• Everything You Need to Know About AI in Customer Service
• What is Data Transformation and How Can it Help ASEAN Unlock AI’s Potential?
• The Sales AI Playbook: 7 Ways AI Can Help You Close More Deals

The post What’s next for ASEAN businesses in 2025? appeared first on Salesforce.

Agentforce allows marketers to forge deeper relationships, from personalised conversations to proactive retention strategies. Let’s explore five use cases that you could build an agent to solve.

1. Automate personalised, two-way conversations on WhatsApp

Interacting 1:1 with every customer who might respond to a promotional campaign on WhatsApp is a tall order for any marketing team. Historically, there’s been limited chat bot capabilities that typically use only keywords, or no ability to respond to an WhatsApp message at all. But with Agentforce, you can connect a customer-facing, autonomous agent that acts like a personal concierge, right from the WhatsApp thread itself. 

Imagine a customer receiving an exclusive offer for a product they’ve shown interest in. They reply with a question, and an agent instantly responds with tailored product recommendations, current offers, or even information on complementary products. If a customer decides to make a purchase, the agent can guide customers through the entire checkout process, from completing the transaction to sending order status and updates. If they need additional help, the agent smoothly transfers the conversation to a human service rep, maintaining continuity.

Why it matters: This approach not only boosts conversions by keeping customers engaged, but it also builds satisfaction and loyalty through timely, relevant responses. By reducing support team load, Agentforce helps deliver a consistent brand experience that feels truly personal and seamless.

Deliver seamless experiences with Agentforce

Automate two-way conversations across any customer interaction.

See how it works

2. Create personalised agendas for event attendees

Events are an invaluable tool for marketers to foster connections and create value for customers. However, ensuring each attendee finds the sessions and resources most relevant to them can be challenging. Trusting the attendees can find the sessions most relevant to them with a self-guided experience can result in more opportunities to abandon registration. Agentforce for Marketing can help you greet visitors on your event website, scaling the ability to provide 1:1 assistance to every visitor. 

The autonomous agent can recommend sessions based on interests and help attendees build personalised agendas. By analysing past attendance and engagement data, the agent can curate an agenda that aligns with each attendee’s preferences and priorities, from keynotes to breakout sessions. For repeat attendees, the agent can even suggest new content based on past sessions and complete registration with their customised agenda.

Why it matters: Personalised agendas help create a memorable event experience that leads to attendee satisfaction and loyalty. By making the registration journey seamless and reducing bounce rates, Agentforce supports stronger connections and builds long-term interest in your events.

3. Capture and qualify leads effortlessly on your website

When visitors land on your website, use Agentforce to welcome them with tailored product or service recommendations, offer exclusive content, and capture contact information. Depending on visitor behaviour, the agent can suggest gated assets such as case studies, demo videos, or even register someone for an event or webinar. This helps avoid self-guided experiences for the visitor, often resulting in opportunities to bounce before becoming a lead or talking to a sales rep.

For highly engaged prospects, the agent can schedule follow-up meetings with sales reps, ensuring they receive immediate value without friction. This proactive, conversational approach nurtures leads by guiding them to the right solutions based on their needs and interests.

Why it matters: Automated lead capture accelerates qualification and boosts conversions. By keeping visitors engaged and moving seamlessly through the pipeline, Agentforce for Marketing nurtures leads naturally, resulting in higher-quality opportunities, and, ultimately, loyal customers.

4. Improve customer journeys with intelligent reprioritisation

Keeping your audience engaged while respecting their communication preferences can be a balancing act. Before customers reach their communication limit, reassign them to the most relevant journey based on both their interests and your business goals. By reprioritising messaging, marketers can ensure high-value content is delivered at the right time, without overwhelming customers. For example, if a customer is close to their communications cap, use Agentforce for Marketing  to prioritise an upcoming product announcement or a VIP event invitation, aligning customer interests with timely opportunities for engagement.

Why it matters: Intelligent reprioritisation leads to higher conversion rates, lower unsubscribe rates, and stronger customer relationships. By adapting dynamically to customer behaviour, Agentforce helps ensure every interaction is relevant so customers stay engaged with your brand without feeling overwhelmed.

5. Reduce churn with proactive, personalised promotions

Retaining customers is just as important as acquiring new ones, especially when they may be at risk of churning. Agents can identify these at-risk customers by analysing specified churn indicators — such as low engagement or decreased purchase frequency — and autonomously send a tailored promotion based on their profile and guardrails set by the marketer.

For example, if a customer’s churn score nears a threshold, reach out with an exclusive loyalty discount or renewal incentive that guides them through the redemption process. By reaching out proactively, agents helps identify where to rekindle interest and engagement before the customer decides to leave.

Why it matters: Proactive AI-powered retention strategies boost customer lifetime value, reduce churn, and foster brand loyalty. With minimal manual effort, Agentforce helps marketers stay connected to customers who might otherwise drift away.

Agentforce is more than an automation or co-pilot tool — it’s an essential partner that empowers you to deliver the seamless, personalised experiences your customers expect. By building agents to implement these five use cases, marketers can increase customer engagement, increase conversions, and build deeper, lasting relationships.

Say hello to Agentforce

Scale your workforce and handle any business use case. Build and customise autonomous AI agents to support your employees and customers 24/7.

Get started now

The post 5 Ways Agentforce Can Elevate Your Marketing Strategy appeared first on Salesforce.

As customers expect faster, more personal, and helpful service, businesses must invest in AI education to empower their teams with the skills and knowledge to harness this technology effectively. Upskilling teams in both CX strategies and AI tools is key to staying competitive and meeting these new demands. 

Why now is the time for AI in customer service

AI isn’t just a trend — it’s changing the way companies help customers. It can help teams solve problems faster, make better recommendations, and give customers what they need before they have to ask. By implementing AI into your customer experience strategy, you can future-proof your team and help your organisation stay competitive.

How AI is already changing customer success

• Personalised service: New insights into customer data can help predict what each person needs and suggest products or solutions that fit them best.
• 24/7 help: AI agents can answer questions any time of day, freeing up your team to work on the more complex interactions.
• Improved customer feedback: Analysis of real-time customer feedback helps teams know what’s working and what’s not, making it easier to improve experiences as they happen.
• Easier order tracking and returns: Customers can track their orders and return items seamlessly and without human involvement.
• Better pricing: Dynamic pricing can help meet the needs of each customer, giving them tailored deals while accounting for market changes.
• Fast problem solving: Companies can identify and fix bugs, mistakes, or problems before customers notice them. 
• VIP service for loyal customers: AI can identify top customers and make sure they get VIP treatment and professional service.
• Smarter interactions: Companies can gauge the opinions of an individual customer using individual sentiment analysis. This enables companies to better understand customers’ emotions, leading to more thoughtful and timely responses and better customer service.

Deploy AI agents with confidence

Provide specialised, always-on support to employees or customers, 24/7.

Get started here

AI education for exceptional CX

Getting started with AI means reskilling your team to feel confident in using it. This process takes time, but it’s worth the effort. And as your team’s AI literacy grows, they’ll start identifying new ways to use the technology. 

Here’s how to build an effective training program for AI skills and encourage employee engagement.

Assess current skills

Identify knowledge gaps, skill gaps, and areas for improvement. Make sure you address core AI literacy and technical skills as well as soft skills that support success with AI tools.

AI literacy and technical skills:

• Foundational AI knowledge: Help your team understand the basics of AI, like machine learning, natural language processing (NLP), and predictive analytics.
• Data analysis: Train your team, as they become more familiar with AI, to understand AI insights and use data to make smart decisions.
• AI tool proficiency: Offer hands-on training with AI-powered CRM systems, chatbots, and other tools they’ll use in their daily routine.
• AI-driven customer journey mapping: Show them how to use AI to map customer journeys and improve personalisation for better customer experiences.
• Customer data privacy and security: Ensure your team is equipped to handle customer data securely and responsibly while benefiting from AI insights.

Soft skills: 

• Critical thinking: Encourage critical thinking, so your team can evaluate AI outputs and make informed decisions when needed.
• Adaptability: Create a culture that embraces learning and tech adoption to keep pace with evolving AI.
• Emotional intelligence: Stress the importance of empathy and the human touch in customer interactions, as these skills complement AI capabilities. It’s all about balance. 
• Collaboration, not replacement: Train your team on when to rely on automation and when a human touch makes all the difference.

By empowering your team with these skills, you’re preparing them for the future of AI. They’ll be confident, skilled, and ready to deliver awesome customer experiences every step of the way.

Develop tailored training

Create personalised learning paths to address the range of AI skills required for your organisation and where team members currently stand. This will help your team focus on what’s relevant to them, speed up improvements, and secure buy-in as team members see themselves in your approach to AI adoption. Hold regular check-ins to assess progress and adjust what topics and resources come next.

AI education in action  

Give your team the opportunity to practice working with AI systems in controlled environments. Mix workshops, online courses, and real-world applications to cement their knowledge in different contexts. This is an ongoing effort, giving your team direct experience as new tools, features, and methods prove to deliver more value. Free hands-on AI courses and certifications are helpful resources your teams can easily access to broaden their exposure to AI.

Encourage collaboration

Share knowledge across teams and departments. A collaborative AI approach ensures that your team adopts best practices company-wide. Encourage team members to proactively share their insights and experiences using new AI tools and skills. This way, different people can try out various options and keep everyone aligned and informed on what’s working and what’s not. 

Reward participation

Recognise employees who embrace AI in their roles. When team members see the impact of AI on their success, engagement and competence skyrockets. Because AI is rapidly changing, it’s useful to constantly explore new ways of measuring progress and how data is understood. Participation gives your team a stake in your organisation’s success with AI.

Build Your First Agent with Agentforce

Create an Agentforce Service Agent to assist with recommendations and bookings.

Build your first agent

+500 points

Project

Quick Start: Build Your First Agent with Agentforce

Continuous learning: The heart of AI success

Getting the most out of AI is all about having a plan for continuous learning. With technology evolving so quickly, what works today might need an upgrade tomorrow. That’s why keeping your CX team in learning mode is essential for staying competitive and delivering great experiences every day. AI isn’t a ‘set it and forget it’ tool. It’s a journey where your team can grow, adapt, and improve.

Encourage your team to stay curious. By exploring new AI tools and techniques, they’ll keep their skills sharp and be ready to jump on the latest innovations as they appear. Regularly refreshing your training programs to reflect these advancements is key. Offer a mix of structured learning, like workshops and online courses, along with informal opportunities for peer learning and collaboration. A culture that champions learning helps your team adapt to new AI-driven solutions faster, so they’re always ready to deliver agile, responsive customer experiences.

And don’t forget to support your team along the way. Whether it’s access to the latest AI tools, time for ongoing training, or the freedom to experiment, weaving continuous learning into your organisation’s AI strategy sets your team up for success. Investing in their growth means keeping them at the forefront of innovation, and equipped to meet your customers’ needs with confidence and expertise.

The future of CX with AI

AI is here to stay, and when used well, it can transform the way you deliver customer experiences. By investing in your team’s AI skills, you’re empowering them to serve customers more effectively, adapt to change with confidence, and lead in a dynamic world.

The post AI Education: How to Reskill Your Team for the Future of Customer Experience appeared first on Salesforce.