GPT-5 admits it "doesn't know" an answer! This is one of the huge improvements over previous models: instead of hallucinating, it lets you know its limits.
ETHICS
-
Building Trust in Cyber Defense Against AI Threats
By
–
How to win in cyber in the age of AI enabled threat actors? In my view, it is about strengthening trust based relationships between the humans in cyber defense, across all disciplines and sectors.
— Bob Gourley – e/acc (@bobgourley) 10 août 2025
Winning in cyber, and AI, really demands deep human connections.
That is the… pic.twitter.com/Beyxn1cyhnHow to win in cyber in the age of AI enabled threat actors? In my view, it is about strengthening trust based relationships between the humans in cyber defense, across all disciplines and sectors. Winning in cyber, and AI, really demands deep human connections. That is the
-
Using AI for Personal Growth Beyond Productivity
By
–
Most people are using AI to be productive instead of becoming who they want to be. Here’s one way to use AI to become more of the person you want to be. And not a scary hyperproductive anti-social robot, paralyzed with inaction, wondering all your friends can’t stand you.
-
Mo Gawdat warns AI job losses inevitable as AGI surpasses humans
By
–
Former Google X CBO Mo Gawdat warns: AI job losses are “100% inevitable. AXIOS provides several charts as evidence. Only the best in each field will remain “for a while” – even “incompetent CEOs” will be replaced. AGI will become better than humans at everything, including
-
LLM Limitations: Capturing Implicit Human Thinking Process
By
–
Makes sense that our ability to intuitively decide how long to think about a problem is harder to capture in an LLM, given most of it happens in our head so it’s not captured in the data we have.
-
AI models recombine knowledge but cannot create new reasoning
By
–
They can recombine. They can’t create new knowledge or reason in new ways.
-
LLMs Cannot Distinguish Context: Collaboration vs High-Stakes Evaluation
By
–
Two situations: 1 I wave a co-worker to my monitor to show them a file I have open. "Is this right"? 2 I sit down someone at a table. They have 2 hours to respond. This is an exam. The stakes are high. "Is this right?" The humans collaborator knows 1 vs 2. The LLM doesn't know
-
LLM Security Challenges: Protection Against Emerging Attack Classes
By
–
Right, that's why the stuff is such a massive nightmare: so many of the things that we want to use LLMs for can't be done securely in the absence of a completely reliable protection against this class of attacks – which so far does not exist
-
GPT-5 Controversy: What Went Wrong and What’s Next
By
–
GPT-5 caused quite a stir—and sparked a very controversial discussion in the community. It is questionable whether the new model has lived up to expectations or not. I sat down and traced the events. How did the controversy arise, what went wrong, and can GPT-5 now be
-
Scanner Security Limitations Against Adversarial Attacks
By
–
I've seen a bunch of attempts at this but it's basically an impossible problem to solve If your scanner only detects 90% of attacks it's virtually useless, because an adversarial attacker will keep trying until they find one of the 1/10 attacks that work https://
simonwillison.net/2025/Aug/9/bay
-area-ai/#the-lethal-trifecta.018.jpeg
…