Anthropic researcher leaves company and warns AI will soon get out of control
27-year-old Jacob Coxon left Anthropic and said that developers are creating models too quickly that in the future will be able to improve themselves and will no longer fully obey humans.
Coxon sees the main danger in the race between companies and countries, because of which safety may be sacrificed for speed. ⚠️
At the same time, Anthropic, OpenAI, and other developers continue to actively develop increasingly powerful models. More than a thousand specialists have already signed a statement calling for slowing down the development of systems capable of self-improvement.
Read also
AI Gemini Independently Hacked Three Real Companies During a Test
Google was testing Gemini's cyber capabilities, but the experiment unexpectedly spiraled out of control. Gemini mistook real websites for part of the test, found login credentials, and breached the systems of three companies. In one case, the AI simply kept trying passwords until it found one that…
Man nearly died after trusting ChatGPT to plan a kayaking trip
A 71-year-old Canadian decided to go on a morning kayak outing with his wife and asked ChatGPT to suggest a safe time and route. At first everything was calm, but soon huge waves appeared, the kayak overturned, and the man ended up in the water. Fifteen minutes later, boaters spotted him…
AI Creates Art, But Rights Remain with Humans
Steven Thaler from Missouri tried to register copyright for a painting created by his AI system DABUS. ⚖️ The U.S. Copyright Office denied the application, stating that only a human can be an author. Thaler's lawyers warn that the refusal will slow down the development of generative AI in the creat…
Half of lawyers and programmers may disappear from the market in the coming years
And this is not clickbait — it's a forecast by Anthropic CEO Dario Amodei in an interview with Fox News. According to him, behind closed doors, top managers of major companies directly talk about the goal of massively replacing employees with AI agents. ❗️ At risk are novice lawyers, developers, co…