Bill Gates warned on an NBC programme that the misuse of AI could result in the deaths of up to 1 billion people. This has been followed by calls for safety measures to be put in place for AI development, as well as a series of departures and moves by researchers at major companies.
Amid growing concerns over AI safety, the Financial Times has reported that more than 12 cases of researchers leaving their jobs have emerged at OpenAI, Anthropic and Google DeepMind, among others. The debate centres on safety and the speed of technological development. The UK government's AI Safety Institute has also reportedly seen employees take leave for reasons including stress and counselling.
Robert O'Callahan, who was responsible for AI chip design tools at Google DeepMind, left the company on the 24th. He cited the view that “AI is developing rapidly” and said that research accelerating this process is not necessarily good for people.
Some are seeking to continue safety research outside companies. Josh Engels, formerly of DeepMind, moved to the independent evaluation organisation METR. Engels expressed concern about the risk of harm within the next 5 years.
At the beginning of this month, Jacob Coxon, who left Anthropic, described the race to develop superintelligence as “gambling with our lives” and said that AI researchers were also concerned that the next 1–2 years would be decisive for safety.
In an interview on an NBC programme, Bill Gates pointed out that malicious actors could combine and use the latest AI tools to create a powerful weapon. He called for governments to put safety measures in place.