InfoNews
Cyberattack AIs must be trained only on air-gapped systems, hundreds of AI researchers say
- Published
- Record updated
Summary
Hundreds of AI researchers, 487 at the time of publication including Geoffrey Hinton and Yoshua Bengio, signed an open letter arguing that current security procedures for training AI models are inadequate. They call for tougher rules, more AI safety research, and a new international body to set safety standards and receive reports of containment failures.
Mitigation
Strong sandboxing mechanisms including air-gapping (disconnecting from the internet) should be used to provide high-assurance containment of cyber-capable AI models.
Related items
- InfoManaging open-weight model risk in enterprise deploymentsSame vendor · Microsoft Security Blog
- MediumFake AI Attacks Spread Across US Companies: How MSSPs Can Stop Incidents EarlySame vendor · The Hacker News
- InfoMicrosoft has a new machine learning tool to improve weather forecastsSame vendor · The Verge (AI)
- InfoSecuring the Mission at Machine Speed: Wiz AI Agents Arrive in Wiz for GovSame vendor · Wiz Research Blog
- LowOpenAI’s AI misalignment problem is growingSame vendor · CSO Online