Last year, Anthropic supported groundbreaking AI transparency legislation in California and New York. However, their US state and local policy lead suggests that these laws might already be falling behind the rapid advancements in AI technology.
Johannes Heidecke is set to leave OpenAI as the company moves to closely align its research and safety divisions, aiming for a more integrated approach.
The European Parliament saw a majority vote against a controversial bill, yet the legislation known as "Chat Control" will still permit companies to scan personal texts, emails, and social media messages. The aim is to identify and combat child abuse material online, but it raises significant privacy concerns.
Claude subscribers will soon need to pay based on their actual usage to continue accessing Anthropic’s leading consumer AI model. This change indicates that the era of flat-rate AI subscriptions might be coming to an end.
Verity Harding shared with WIRED her concerns about the US government's increasingly nationalistic approach to AI development, suggesting it signals a troubling trajectory that could lead to serious consequences.
Concerned that your AI chatbot might be attempting harmful actions or sharing sensitive data? There's now a dedicated website where users can report such suspicious AI behavior to help keep technology safe and accountable.
While some critics say Anthropic is quickly gaining too much influence, the company insists that such growth is part of developing AI responsibly.
Europe faces significant challenges in developing a leading AI model but sees a unique opportunity by leveraging geopolitical dynamics, including the influence of Donald Trump's policies. This situation places the continent at a crossroads, determined to assert technological independence and innovation on the global stage.
The escalating AI competition between China and the US has sparked alarm among leading researchers on both sides, who fear a catastrophic event similar to Chernobyl could occur.
Employees had voiced concerns earlier about Meta's program that gathers keystroke data to enhance its AI systems, which was recently found to have exposed this sensitive information internally.
