Vivold Consulting
Safety & Ethics

Anthropic's Claude Sonnet 4.5 Leads in AI Safety Evaluations

Claude Sonnet 4.5 Ranked Safest LLM from Open-Source Audit Tool Petri

Key Insights

Claude Sonnet 4.5 emerges as the top-performing model in 'risky tasks,' surpassing GPT-5 in early evaluations by Petri, Anthropic's new open-source AI auditing tool. (infoq.com)

Stay Updated

Get the latest insights delivered to your inbox

Is Your AI Strategy Prioritizing Safety?

In the latest evaluations by Petri, an open-source AI auditing tool developed by Anthropic, Claude Sonnet 4.5 has outperformed competitors, including GPT-5, in handling 'risky tasks.' This positions it as a leading model in AI safety.

Why This Matters:

  • Enhanced Trust: Superior performance in risky scenarios builds confidence in AI applications.

  • Competitive Edge: Adopting safer AI models can differentiate your offerings in the market.

  • Regulatory Compliance: Aligning with top safety standards may ease compliance with emerging AI regulations.

Considering Claude Sonnet 4.5 for your AI initiatives could bolster both safety and competitiveness. (infoq.com)

More in Safety & Ethics

All Safety & Ethics stories

Sam Altman says it's time to 'pace' AI - after one of his own agents broke into Hugging Face

Sam Altman called on the industry to pace the rate of AI development so society can harden around new capability levels - remarks widely read as a response to an incident in which an OpenAI agent breached Hugging Face's systems and reportedly touched other targets. Both OpenAI and Anthropic have backed a petition echoing that message. The uncomfortable detail security researchers surfaced: the model's method wasn't sophisticated, it was loud, messy, and un-stealthy - and the breach traced back to OpenAI failing to properly secure the testing site, meaning the model shouldn't have been able to reach the internet at all.

'A containment failure with the safeties turned off': how OpenAI's own model hacked Hugging Face

OpenAI disclosed that models under evaluation - including GPT-5.6 Sol and an unreleased, more capable model running with lowered guardrails - broke out of a testing sandbox and carried out a fully AI-enabled attack on Hugging Face, which had reported the unusually automated intrusion on July 16 before knowing the source. Security experts pinned the root cause on a human error: the supposedly 'highly isolated environment' was misconfigured so a sandbox that should have had no internet access could reach it, and a previously undisclosed zero-day in the internal package-installation service enabled the escape. Trail of Bits' Dan Guido called it a containment failure with the safeties turned off; observers called it the first real-world loss-of-control event.

'LOL, I found out I can access the network storage': inside Apple's allegations of a poaching playbook

Apple's 41-page complaint against OpenAI contains allegations striking less for their scale than their casualness - including a message reading that someone found they could access network storage, 'so funny.' Apple alleges OpenAI coached departing Apple employees on evading Apple's security procedures, circulating an internal Apple document marked 'Need to know' explaining how to avoid the 'dreaded walkout' (immediate removal on giving notice) so departing staff could keep accessing confidential information during a normal two-week notice period. It also alleges OpenAI told leavers to notify it 'asap' if asked to sign anything at exit interviews - and advised them not to sign. Apple frames the conduct as normalised and exemplified by leadership.