Interesting Engineering

China: Single voice command exposes humanoid robots to hijacking and cascading attacks

Chinese security tests show robots hacked in minutes via voice or wireless flaws, spreading attacks to other machines and ...

OpenAI says prompt injections that can trick AI browsers like ChatGPT Atlas may never be fully solved—experts say risks are a feature not a bug

“Prompt injection, much like scams and social engineering on the web, is unlikely to ever be fully ‘solved,'” OpenAI wrote in ...

Your AI browser can be hijacked by prompt injection, OpenAI just patched Atlas

OpenAI says it has patched ChatGPT Atlas after internal red teaming found new prompt injection attacks that can hijack AI ...

OpenAI Says Prompt Injections a Challenge for AI Browsers, Builds an Attacker to Train ChatGPT Atlas

OpenAI says prompt injections remain a key risk for AI browsers and is using an AI attacker to train ChatGPT Atlas.