Cut Through The AI Chaos

If you are still tied to your IDE, whether Cursor or Copilot, you are working a year behind. Coding turned out to be AI’s home territory. The best tooling has moved out of the editor and onto the command line, and the senior engineer’s job is to train the AI, not review its output.

I wrote the previous version of this post in March 2025, updated it once in August, and it has been linked from almost everything I have written about AI engineering since. The fundamentals from that post still hold: keep changes small, build guardrails, document ruthlessly, and make sure every change gets verified before it ships. One thing has had to move with the volume. “Verified” used to mean “read by you”. With modern agent throughput, it has to mean “checked by tests, by type checkers, by automated gates, or by you where your judgement matters”. The check still happens; it just does not always happen in your head.

One distinction has hardened over the last year, and it matters for what follows: vibe coding and agentic engineering are different practices. Vibe coding, where you do not really check the results and just look at the output, is fine unless you are shipping. Agentic engineering is the other thing. It does not mean reading every diff. It means making a measured call about which diffs need your eyes and which do not. I tend to skim UI diffs. I read security and database diffs carefully.

That judgement, about where to spend your attention, is itself a senior skill and one worth getting good at. Every diff still gets checked; not always by you. Tests, type checkers, and automated gates pick up the rest, and you are training the agent so the diffs that do reach your eyes are the ones worth the attention. Conflating the two is how people end up shipping vibe-coded output into production and getting burned.

Anthropic announced the most powerful cyberweapon ever built last week and kept it, granting access to forty companies while the US government got a press release. Harry Truman, who nationalised atomic research inside three years, would not have stood for it.

Project Glasswing, Anthropic’s cyberdefence initiative, restricts Claude Mythos to a named coalition including AWS, Apple, Google, Microsoft, NVIDIA, and JPMorgan Chase, on the grounds that the model is too dangerous for public release.¹ Mythos finds vulnerabilities in every major operating system and browser autonomously, and in doing so it has already discovered a 27-year-old flaw in OpenBSD, a 16-year-old bug in FFmpeg that five million automated tests had missed, and a chain of Linux kernel vulnerabilities that allowed complete privilege escalation, though the details have held up less well under external scrutiny than they did in the press release. Anthropic holds the knowledge of how to find more, and it has decided who gets access to it.

Eight days later Anthropic shipped Claude Opus 4.7 to everybody else, with the cyber capabilities deliberately reduced during training: same lineage, defanged.² The dangerous version goes to the coalition, the distilled version goes to the paying public, and the gap closes as the capability diffuses. Anyone who read Kokotajlo’s AI 2027 forecast last year is watching it play out on schedule.³

Washington has responded by applying for access. Senator Coons called Project Glasswing “a landmark moment”, Senator Warner urged industry to patch faster, and Treasury Department staff ignored a standing ban on Anthropic products to queue up for Mythos access anyway.⁴ The picture is the same across the Atlantic, where the EU AI Act is in force and the UK AI Security Institute is active, but neither treats Glasswing as a challenge to state authority. Engagement, not confrontation, is the pattern on both sides, and it looks nothing like the response Truman gave to atomic research. He did not ask for a subscription.

What if the AI already knew what you were working on? What if it remembered what you decided last week, what your board presentation covers, what your team structure looks like, because it had read your notes?

That is what happens when you put Claude Code in a folder. You do not need to know how to code, and you do not need a plan for what goes in the folder. You just need something on your mind, and about ten minutes.