Google unveils CodeMender AI to rival Anthropic Mythos
SkimNews Take
The rapid expansion of AI into critical infrastructure like cybersecurity is pushing developers to prioritize resilience and reliability, as early outages quickly become market differentiators.
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Google announced at I/O that its CodeMender API is now open for select expert testing and broader external rollout to flag and fix code vulnerabilities.
- Koray Kavukcuoglu (Google DeepMind CTO) said CodeMender aims to “help secure the world’s code bases” by automatically detecting and repairing security gaps.
- Sundar Pichai noted Anthropic’s Mythos showed the value of large models for security and asserted Google can match that capability.
- Anthropic released a preview of its Mythos AI security model, described as too powerful for public release and credited with restoring its standing with the US government after a supply‑chain risk lawsuit.
- OpenAI launched its own code‑security AI offering shortly after Anthropic’s Mythos preview, joining the race for high‑value enterprise security contracts.
Why it matters: Enterprises and governments gain a new AI audit solution, while Anthropic’s Mythos delay and OpenAI’s rival offering heighten competition for lucrative security contracts, pushing all three firms to monetize AI‑driven code protection.
