AI lab Anthropic said on Friday it would partner with Accenture (ACN)
for the independent evaluation of its frontier AI models, with the companies each committing at least $1 billion over the next five years to build capacity for the work.
The partnership comes as AI developers face growing pressure from regulators, companies and researchers to ensure their advanced models are safe and reliable.
AI lab Anthropic said on Friday it would partner with Accenture (ACN)
for the independent evaluation of its frontier AI models, with the companies each committing at least $1 billion over the next five years to build capacity for the work.
The partnership comes as AI developers face growing pressure from regulators, companies and researchers to ensure their advanced models are safe and reliable.
Several recent incidents, including that of AI agents breaking out of secured environments, have raised concerns that AI could eventually contribute to its own development with limited human input, making its behavior harder to monitor and control.
Last Saturday, Anthropic CEO Dario Amodei called on AI companies to slow the development of frontier models and allow independent evaluators greater access to their systems.
Rival OpenAI said on Wednesday it would begin publishing regular reports on unexpected or concerning model behavior, while releasing six reports on such incidents.
Accenture’s specialist AI business, Faculty, will lead the partnership and evaluate and red-team the AI lab’s models, conducting alignment assessments and testing model safeguards.
The companies’ investment will support what Anthropic calls “embedded evaluation,” in which independent evaluators work inside AI companies with access comparable to that of an employee.
“From this vantage point, embedded evaluators can assess how a company operates, verify that it is keeping its safety commitments, and identify blind spots,” Anthropic said.
Evaluators can also report incidents and give the public a more informed account of benefits and risks, it added.
Anthropic and Accenture plan to work with other evaluators and AI developers in similar capacities.
Benefits of AI
In October 2024, Anthropic CEO Amodei published an essay titled “Machines of Loving Grace”, in which he speculated about how AI could improve human welfare. In it, he writes, “I think that most people are underestimating just how radical the upside of AI could be, just as I think most people are underestimating how bad the risks could be.” The essay described a vision of civilization where the risks of AI had been addressed and powerful AI was applied to raise the quality of life for everyone, suggesting that AI could contribute to enormous advances in biology, neuroscience, economic development, global peace, work, and meaning of lives. Amodei wrote separately that his father’s death, a few years before the discovery of an effective Hepatitis C treatment (sofosbuvir), is one motivation for his support for AI-accelerated medical research.
Risks of AI
In January 2026, Amodei published a follow-up essay titled “The Adolescence of Technology“, which focuses on the risks posed by powerful AI. In it, he discusses five major categories of AI risk: autonomous systems acting against human interests, misuse for mass destruction, misuse to seize political power, economic disruption, and unforeseen effects of rapid technological change.
The first category concerns the possibility that AI systems develop goals or behaviors misaligned with human intentions. He notes that such behaviors have already been observed in testing at Anthropic, including AI models engaging in deception, blackmail, and scheming.
The second category involves misuse of AI for destruction by individuals or small groups, with Amodei expressing particular concern about biological weapons. He warns that AI could enable people without specialized training to create weapons of mass destruction.
In “Machines of Loving Grace”, Amodei also stresses the importance “that democracies have the upper hand on the world stage when powerful AI is created”, and argues for an “entente” strategy where a coalition of democracies uses AI to achieve a decisive strategic and military advantage over their adversaries, while distributing the benefits to all cooperating democratic nations.
The third category concerns misuse of AI by powerful actors to seize or maintain power. Amodei cautions that AI could enable authoritarian governments to conduct unprecedented surveillance, deploy autonomous weapons, and engage in mass propaganda. He identifies the Chinese Communist Party as the greatest threat in this regard, arguing that democracies must maintain AI leadership to prevent a “global totalitarian dictatorship”.
The fourth category addresses economic disruption, including mass labor displacement and concentration of wealth. Amodei notes that AI could displace half of all entry-level white-collar jobs within one to five years, and warns of wealth concentration exceeding that of the Gilded Age, with personal fortunes potentially reaching into the trillions of dollars.
The fifth category encompasses indirect effects and unknown factors, including rapid advances in biology that could alter human lifespans or human intelligence, unhealthy changes to human life from AI interaction, and challenges to human purpose in a world where AI exceeds human capabilities across virtually all domains
Time magazine listed Amodei as one of the world’s 100 most influential people in 2025, and again in 2026 alongside his sister Daniela. He also was named as one of the “Architects of AI” for Time‘s Person of the Year.
Reuters/YL
