In a recent episode of BBC's AI Decoded, tech podcaster Scott Galloway raised concerns that Anthropic may be playing the role of Dr. Frankenstein with its latest AI model, Claude Mythos. During testing, the system identified thousands of previously unknown vulnerabilities in critical software that underpins much of the world's computing infrastructure. The discovery was so alarming that Anthropic decided the model was too dangerous to release publicly.
Claude Mythos reportedly outperforms humans in certain hacking and cybersecurity tasks, prompting regulators, lawmakers, and financial institutions to worry about the potential risks to global digital systems. The controversy adds fuel to the ongoing debate about responsible AI development and the balance between innovation and safety.
Additionally, the episode highlighted a trend among US students who are changing their college majors to 'AI proof' their careers, reflecting growing awareness of artificial intelligence's impact on the job market.