Product & Tech Leadership Advisor, Consultant, Coach & Mentor
Tech, Software Development, Science, History, Economics, Politics
https://www.linkedin.com/in/robertbowley
https://pragmaticpartners.co.uk
Manchester, UK

@robbowley.net
Product & Tech Leadership Advisor, Consultant, Coach & Mentor
Tech, Software Development, Science, History, Economics, Politics
https://www.linkedin.com/in/robertbowley
https://pragmaticpartners.co.uk
Manchester, UK
The mechanism of coding was largely solved by the mid-1960s: we knew how to tell a machine to do anything computable. What wasn’t solved, and arguably can’t be, is knowing what to tell it, confirming it’s right and keeping it right as requirements shift.
Margaret Hamilton, trailblazer whose software powered Apollo 11 moon landing, dies at 90
Someone just said to me: "AI is like Jesus. I like him, just not his followers." Serious contendor for quote of the decade
I've published detailed notes and an annotated transcript to accompany the video of the keynote I gave at @wearedevelopers.bsky.social World Congress North America in San Jose on Friday - here's my rundown of everything that's happened with LLMs in 2026 so far simonwillison.net/2026/Sep/27/...
A good reminder to (re)read No Silver Bullet. Tools shrink accidental complexity; essential complexity remains.
I made it into Martin Fowler's Fragments again 😍
If your primary objective with AI is anything but learning faster, you're spending a lot of time optimising the wrong thing. Progress moves at the speed of its slowest step. For knowledge work, that step is understanding. It always has been.
Learning how to use Ableton (music creation software). It is like being in the cockpit of a space shuttle
There are always ways to jailbreak the safegaurds in the LLMs and cost yes I called that out in my article, but it will get cheaper e.g. Deepseak and also OS models getting better over time
which is also evidenced on zerodayclock.com
exploiting software flaws is now the top way attackers get in (31%, up from 20%), with AI cutting the time to exploit from months to hours.
www.verizon.com/about/news/b...
The UK's NCSC says AI will "almost certainly" increase the frequency and intensity of cyber threats up to 2027.
Getting busier with busywork example of the week
Thx for your presentation deck btw sources were useful for my blog I think I cited at least one
On the increase in cyber attacks
Lol the anger not just me then. Get you re LLMs making it hard to do bad things (unless guardrails intentionally removed e.g. huggingface) but I don't think will be long b4 won't take much to build an LLM without any and just use more brute force
Also orgs are getting fast and loose wiring agents into their systems as fast as they can, chasing promised productivity, building on something we don’t know how to contain, and shipping it to everyone while we work it out.
At least it's out of my head now, but I'm genuinely worried, particularly for the tsunami of cyber attacks I see coming
It has all been making me really angry – the media lapping it up and letting it go unchallenged. So angry I’ve been struggling to be coherent in collecting my thoughts on it all. Every time I think I’m there another sensational headline or take comes up and I’m flipping tables in my head again.