Anthropic and Accenture have announced a partnership aimed at embedding teams within the Claude developer to evaluate and assess the safety of advanced AI models.
The companies expect to commit at least $1bn each over the next five years with the investment directed towards building capacity in AI safety.
Go deeper with GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
The collaboration will see Accenture, through its acquired AI subsidiary Faculty, work alongside Anthropic’s staff and other safety partners.
The focus will be on evaluating and “red-teaming” AI models, conducting alignment assessments, and testing the effectiveness of existing model safeguards.
According to the companies, embedded evaluators working within AI developers represents a newer approach in the field, distinct from traditional external evaluation methods.
Accenture’s involvement leverages experience gained through Faculty, which has previously worked with governments, defence bodies, and healthcare systems. This includes developing the UK National Health Service’s (NHS) Early Warning System during the COVID-19 pandemic.
Faculty’s role in the project is to use its expertise in building and evaluating complex AI systems.
Accenture chief technology officer and Faculty CEO Marc Warner said: “Faculty, which is now part of Accenture, was founded on the belief that AI should be safe by design, not safe by accident. That conviction has guided our work with governments, defence, and healthcare systems across the world.
“Joining forces with Anthropic as embedded evaluators is exactly the kind of work Faculty was built to do—and as part of Accenture, we have the platform to do it at a scale that can genuinely move the industry.”
The companies described the new approach as allowing “embedded evaluators” to operate with access similar to employees, monitoring how models are trained and deployed, assessing internal processes, and verifying safety commitments. However, they caution that standards for embedded evaluation are not yet fully defined.
Anthropic stated that it will directly fund Accenture’s participation in the initial phase, while also exploring partnerships with other evaluators and nonprofit organisations.
Accenture chair and CEO Julie Sweet said: “Accenture is bringing together a dedicated team with deep AI, security and industry expertise to work alongside Anthropic. Safety requires both deep technical expertise and a clear understanding of how AI is used in the real world.
“Embedded evaluation is an emerging area and we look forward to partnering with Anthropic to help accelerate the development of embedded evaluators, which we see as an important part of the safety landscape going forward.”
The collaboration is non-exclusive, and both companies anticipate working with other partners and evaluators as the field develops.
Earlier this month, Accenture and Google Cloud launched the Accenture Gemini Enterprise Business Group, a global initiative focused on helping companies deploy and expand agentic AI and data solutions.
