OpenAI says AI browsers like ChatGPT Atlas may never be fully secure from hackers—and experts say the risks are ‘a feature not a bug’

OpenAI warns that AI browsers like ChatGPT Atlas can never be fully secure against hackers. Experts suggest cybersecurity risks are inherent to AI technology, making complete protection virtually impossible. The company acknowledges ongoing vulnerability challenges.