By Kevin Shepherdson, Founder & CEO, Straits interactive
Each year, Straits Interactive publishes predictions of trends that organisations in the region should monitor. In 2024, we combined privacy and generative AI trends. For 2025, we are focusing exclusively on generative AI trends in the workplace, highlighting the implications for governance, risk management, and compliance (GRC) professionals, including Data Protection Officers (DPOs).
Generative AI is set to redefine workplace dynamics in 2025, driving productivity, innovation, and efficiency across industries. However, alongside these opportunities come significant privacy, ethical, and operational risks that must be carefully managed.
In this article, our research team has identified six key trends in the workplace—the 6 Cs: Collection of Data, Compute Power, Context Window, Chain of Thought, Customisation, and Control—that are critical for harnessing AI's potential responsibly, especially for SMEs. We reflect on developments from 2024 and anticipate their evolution in 2025, focusing on practical implications for organisations and GRC professionals.
1. Collection of Data: Harnessing Internal Knowledge
Workplace Context
In 2025, the trend toward leveraging internal knowledge repositories will reshape how organisations approach AI training. Companies will increasingly utilise customer interactions, policy documents, operational manuals, and archived communications to build AI models that reflect their unique expertise. This shift addresses the growing demand for contextually relevant AI outputs while also raising critical issues around data governance, copyright, and ethical data use.
Additionally, the rise of Small Language Models (SLMs) will allow organisations to develop and deploy AI systems trained solely on internal, proprietary data, reducing dependency on external datasets and mitigating risks related to copyright violations and data sovereignty. SLMs also provide enhanced privacy protection and are computationally more efficient, making them an attractive alternative for enterprises managing sensitive data.
2024 Developments
1. Retrieval-Augmented Generation (RAG): Organisations increasingly adopted RAG techniques, enabling AI models to extract insights from proprietary datasets, thereby providing more contextualised and relevant outputs.
2. Data Privacy Concerns: The rise in internal data utilisation heightened awareness of potential biases and the inadvertent inclusion of personally identifiable information (PII), leading to stricter data governance measures.
2025 Trends
1. Enhanced Data Governance: Organisations will implement robust frameworks to ensure ethical data collection and processing, mitigating risks associated with bias, privacy breaches, and copyright issues. More governance and regulatory frameworks will be introduced to address these challenges.
2. Integration of External Data: Combining internal data with curated external sources will provide a more comprehensive knowledge base, enhancing AI model accuracy.
3. Adoption of Small Language Models (SLMs): Organisations will increasingly leverage SLMs to maintain better control over their data while ensuring efficient and cost-effective AI applications.
Implications for Organisations
1. Improved Decision-Making: Leveraging internal knowledge bases will enable faster and more informed workplace decisions.
2. Ethical Compliance: Implementing stringent data governance practices will ensure compliance with privacy regulations and ethical standards.
3. Bias Mitigation: Regular audits and diverse data sourcing will help in identifying and correcting biases in AI outputs.
2. Compute Power: Enabling Scalable Workplace AI
Workplace Context
The demand for scalable, lower-cost, and energy-efficient AI solutions will accelerate in 2025, driven by innovations in compute power. Organisations will benefit from advances such as NVIDIA’s Blackwell architecture and cloud-based AI platforms, enabling widespread adoption of sophisticated AI technologies. This trend will make cutting-edge AI more accessible to SMEs while aligning with sustainability goals and improving operational efficiency.
2024 Developments
1. Amazon’s Ultracluster Supercomputer and Trainium Chips: These innovations made scalable and sustainable AI compute resources more accessible.
2. NVIDIA’s Blackwell Architecture: The introduction of the GB200 Grace Blackwell Superchip enhanced AI tasks and supported large language models with significant performance improvements in AI training and inference.
2025 Trends
1. Cloud-Based AI: Hybrid cloud environments will scale AI solutions across organisations.
2. Energy Efficiency: Compute resources will prioritise sustainability.
3. Democratised Access: Scalable AI platforms will make advanced technologies accessible to SMEs.
Implications for Organisations
1. Cost Efficiency: Cloud-based compute reduces upfront investments.
2. Sustainability Goals: Aligning AI operations with environmental, social, and governance (ESG) targets.
3. Workplace Transformation: Affordable compute power empowers smaller teams to access cutting-edge AI tools.
3. Context Window: Enhancing Collaboration and Communication
Workplace Context
In 2025, advancements in AI's context window capabilities will redefine workplace collaboration. The trend toward multi-session or even infinite memory will allow AI systems to retain and utilise context across multiple interactions, enabling seamless teamwork. By integrating larger and more complex datasets, organisations can unlock new levels of productivity and continuity in their projects.
2024 Developments
1. Google’s Gemini Models: These models introduced million-token context windows, while advances in multimodal processing allowed integration of text, images, and other data forms.
2025 Trends
1. Multi-Session / Infinite Memory: AI systems will develop capabilities to retain context across multiple interactions, effectively achieving an “infinite memory.”
2. Knowledge Integration: Combining multimodal inputs for comprehensive project support.
3. Personalised Assistance: Context-aware virtual assistants providing tailored solutions.
Implications for Organisations
1. Streamlined Communication: AI reduces miscommunication by summarising conversations.
2. Productivity Gains: AI refines complex documents efficiently.
3. Seamless Collaboration: Projects retain continuity across team transitions.
4. Chain of Thought: AI-Powered Workplace Reasoning
Workplace Context
Chain-of-thought reasoning is poised to gain prominence in 2025 as a critical workplace competency. This advanced prompting method enables AI to explain its outputs step-by-step, enhancing problem-solving and decision-making. From strategic planning to compliance workflows, organisations will increasingly rely on AI tools to tackle complex tasks with greater transparency and accountability.
2024 Developments
1. OpenAI’s O3 Model: Inspired by human reasoning, the O-series models improved multi-step problem-solving, while autonomous AI agents began executing complex tasks with minimal human intervention.
2025 Trends
1. Problem-Solving Support: AI tools to guide employees through multi-step workflows.
2. Scenario Analysis: Predictive analytics to inform workplace decision-making.
3. Explainable Outputs: Transparent reasoning processes for accountability.
Implications for Organisations
1. Enhanced Decision Support: AI as an advisor for critical decisions.
2. Employee Empowerment: Transparent outputs building trust in AI-generated recommendations.
3. Operational Efficiency: Automating reasoning-heavy tasks to free up employee time.
5. Customisation: Harnessing AI and Autonomous Agents for Workplace Transformation
Workplace Context
The proliferation of AI agents in 2025 will drive a new wave of workplace customisation. These autonomous systems will be tailored to specific industries, departments, and roles, enabling organisations to automate repetitive tasks, deliver personalised services, and optimise decision-making processes. This trend will enhance productivity and innovation by addressing unique organisational needs. The role of a Chief AI Officer (CAIO) is also emerging as organisations seek leadership in AI strategy and governance, ensuring alignment with ethical and business objectives.
2024 Developments
1. No-Code Platforms: Democratised AI customisation, enabling non-experts to adapt models.
2. AI Agents: Integrated into workflows, automating repetitive tasks like scheduling and customer support.
2025 Trends
1. Industry-Specific AI: Tailored solutions for unique workflows in sectors like healthcare and retail.
2. Proactive AI Agents: Predictive capabilities supporting decision-making.
3. Employee-Driven AI: Empowering employees to design and deploy AI agents for personalised use.
Implications for Organisations
1. Workforce Enablement: Encouraging innovation through employee-driven AI customisation.
2. Faster ROI: Addressing immediate organisational needs with tailored AI tools.
3. Competitive Edge: Solving unique challenges and enhancing productivity with customised agents.
6. Control: Ensuring Ethical and Effective AI Deployment
Workplace Context
Finally, as AI systems take on increasingly agentic tasks in 2025, the need for robust control mechanisms will intensify. Organisations must establish comprehensive frameworks to monitor AI operations and align outputs with ethical standards and organisational values. This trend underscores the importance of embedding data and AI governance within organisations to mitigate risks, ensure compliance, and build trust in AI systems.
With the rising complexity of AI applications, the Chief AI Officer (CAIO) will play a vital role in ensuring AI creates organisational value, maintaining transparency and accountability in AI deployment while working closely with the AI governance team in AI ethics, risk management, and regulatory compliance.
2024 Developments
1. EU AI Act: Mandated transparency and accountability, setting global benchmarks for ethical AI.
2. FTC Oversight: Highlighted the importance of governance by requiring companies to delete AI models trained on improperly sourced data.
2025 Trends
1. Agentic AI Tasks: Automating complex workflows like policy drafting and resource scheduling.
2. Ethical Oversight: Advanced systems monitoring AI operations for compliance.
3. AI Accountability: Real-time tracking tools mitigating risks and ensuring traceability.
Implications for Organisations
1. Trust Building: Transparent mechanisms enhancing trust among stakeholders.
2. Safe Autonomy: Robust oversight enabling AI to handle more tasks responsibly.
3. Regulatory Compliance: Avoiding legal and reputational risks through strong governance.
Key Takeaways for GRC Professionals
1. Data Governance: Establish frameworks for ethical data collection, storage, and processing, with regular audits to mitigate risks such as bias, PII leakage, and non-compliance.
2. Ethical Practices: Adopt energy-efficient and privacy-conscious compute resources.
3. Context Management: Develop clear policies for extended context usage.
4. Explainable AI: Advocate for transparent reasoning in AI systems.
5. Customisation Compliance: Prevent biased outputs and align with data protection regulations.
6. Control Mechanisms: Implement real-time monitoring for agentic AI tasks
My Final Thoughts
GRC professionals and DPOs play a critical role in aligning generative AI with ethical, regulatory, and data privacy goals. By prioritising robust governance, transparency, and compliance, organisations can ensure generative AI becomes a responsible and transformative force in 2025.
I highly encourage readers to attend our joint Generative AI courses with the Singapore Management University Academy to gain deeper insights into these emerging trends. In 2025, we look forward to aligning our Capabara Generative AI platform with the 6 Cs to empower our clients with enhanced generative AI capabilities.
