AI COMPANIES MAKE POWERFUL TECH – BUT THEY'RE ALSO SAVVY MARKETERS

Summary from the TLDR Marketing newsletter

Anthropic's release of Claude Mythos, a cybersecurity-focused AI model, has generated significant attention partly because the company withheld it from public access and called it a "reckoning" for the industry. Cybersecurity experts are already pushing back, with one noting that some of the claimed zero-day vulnerability discoveries are less significant to real-world security than the messaging suggests. The withholding strategy creates demand, a tactic previously used by OpenAI when it delayed the release of GPT-2 and its video generator Sora.