Signal File: AI gets grounded in real work
From reducing everyday workload to reshaping talent strategies and system design, AI is moving to practical applications, where trust, usability and workforce realities define its value
AI is being deployed into workplaces that are already strained, shifting the focus to where it genuinely reduces effort, where it creates new friction, and where it exposes gaps in how work is organised.
This week’s signals point to a clear pattern. People are looking for relief from low-value work. Organisations are testing how systems connect. And talent strategies are beginning to shift around who can navigate this complexity. The result is a more grounded phase of adoption, where AI is judged by how well it fits into the realities of work.
What workers want from AI is less grind
New research from Anthropic, based on interviews with more than 80,000 people across 150 countries, finds that people primarily want AI to ease technical and repetitive work so they can focus on higher-value thinking. Their biggest concerns are unreliability, job disruption and the risk of losing human agency, suggesting demand is being shaped as much by workplace strain as by technological ambition.
In action: The strongest AI use cases will remove friction. Tools that reduce drain while preserving human control are more likely to earn trust.
AI systems begin working across models
Microsoft is introducing Copilot features that allow users to run multiple AI models simultaneously within the same workflow, combining outputs from systems like OpenAI’s GPT and Anthropic’s Claude. In its new ‘Critique’ function, one model generates a response while another reviews it for accuracy and quality, signalling a move away from single-model reliance toward coordinated AI systems.
In action: Design for multi-model workflows. As AI tools begin to work together, organisations should think in systems, combining models for generation, validation and decision support rather than relying on one source of output.
Women over 50 emerge as overlooked AI talent
Women over 50 are increasingly being recognised as a valuable yet underutilised workforce segment. Their experience, adaptability and resilience position them well for navigating the complexity and constant change of an AI-driven economy, yet they remain underrepresented in hiring and workforce strategies.
In action: As AI reshapes work, organisations should look beyond traditional hiring patterns to unlock experienced, adaptable talent already in the workforce.
Enterprise AI adoption begins to shift toward Anthropic
New data from Ramp shows that nearly three-quarters of businesses buying AI for the first time are choosing Anthropic over OpenAI, marking a reversal from late 2025. The shift comes as OpenAI refocuses on coding and enterprise use cases, signalling how quickly preferences are evolving as the market matures.
In action: As the AI landscape fragments, organisations should prioritise flexibility and interoperability over early standardisation.
