The researchers argue that CoT monitoring can help researchers detect when models begin to exploit flaws in their training, manipulate data, or fall victim to malicious user manipulation. Any issues that are found can then either be “blocked, or replaced with safer actions, or reviewed in more depth.”

Mark Zuckerberg says Meta is building out some massive data centers to power its AI ambitions and compete with OpenAI, Google, and Anthropic.

Monitoring AI's train of thought is critical for improving AI safety and catching deception. But we're at risk of losing this ability.

As tech giants like Microsoft, IBM, and Google shed thousands of workers, experts weigh in on the motivations behind these layoffs.

No AI company scored better than “weak” in SaferAI’s assessment of their risk management maturity. The highest scorer was Anthropic (35%), followed by OpenAI (33%), Meta (22%), and Google DeepMind (20%). Elon Musk’s xAI scored 18%.

Baird analyst Colin Sebastian joins the GeekWire Podcast to discuss Amazon, Google, and Meta’s AI strategies, tech investing trends, and how today’s AI boom compares to the dot-com era.

It's not just the spending that will be big. Meta's Mark Zuckerberg envisions a single data center, the footprint of which would occupy as much acreage as a good chunk of Manhattan.

Google, Meta, and CoreWeave lead an expected $90 billion wave of AI and energy investments in Pennsylvania, marking a shift toward infrastructure-driven AI growth.