Anthropic · Safety and security
Anthropic and Accenture each put $1 billion over five years into independent evaluation
Under the partnership announced on September 18, evaluators will examine models from inside the company rather than from outside: their access will be comparable to an employee’s, letting them observe training and follow development. Faculty, Accenture’s AI arm, will run the work, covering red-teaming, alignment assessments and tests of model safeguards. The deal is non-exclusive, and Anthropic is also in talks with METR and other nonprofit evaluators. The contested part: the company being evaluated is paying for it.
Sources
- Anthropic, “Partnering with Accenture on embedded evaluation”, (anthropic.com)
- Accenture, “Accenture and Anthropic Partner to Build Team of Embedded Evaluators at Anthropic”, (newsroom.accenture.com)
About this story
This story was posted on Instagram by @jarrus.tech on Sept. 22, 2026.
Spotted an error in this story? [email protected] · Instagram
Short link: thejarrus.com/en/h7-34
This story in Turkish: Anthropic ve Accenture, bağımsız değerlendirme için beş yılda 1’er milyar dolar ayırdı
On the same topic
Anthropic releases Claude Haiku 5.5 at 90% lower prices
Claude Haiku 5.5, released Oct. 7, costs 90% less than Haiku 4.5 for prompts up to 100,000 tokens. It leads on benchmarks but uses more tokens per task.
A developer used Claude Opus 5.5 to build a 3D warehouse management system
In WareTrack, which Dilum Sanjaya built with Claude Opus 5.5, trucks, forklifts, inventory and shipments live in one 3D view that plays like a strategy game.
The week in AI: Sept. 28-Oct. 3, 2026
The week’s top AI stories, Sept. 28-Oct. 3: Gemini 4 Argon, OpenAI’s dots agents, Tavus Griffin, an FTC probe and an AI consciousness debate at the Vatican.