Are you testing a model or a harness? People conflate the two.
Can't you do that with regular Unix permissions?
Age eleven and had access to a chemistry set that a relative gifted. It had sulfur, but the saltpeter, and charcoal came from elsewhere. The 1960s encyclopedia had the instructions.
Let the kids play.
No. Excel is a general purpose tool that can be used for calculating tasks that are good, neutral, or evil things. It's a fancy calculator.
Looks good for light-duty uses. Scared for my fingers for anything heavy.
Trillionaire
If Ghengis Khan was raping & pillaging in 2026 his Wikipedia article might have a more modern slant.
They become hyperstition engines.
When we do these it's a fine-tuned classifier, generally a BERT class model. Works quite well when you sanitize input and output with low latency/cost.
Even if you spend a crazy amount of money with them for advertising they will try to gaslight you if they make a mistake.
Finnish if you want to go hard mode.
I’ve worked with their consultants and they were lovely. They hate Azure too.
Sat across from Dr. Ruth on the Acela going from NYC to DC for business. She was lovely and sharp as ever.
Like video gaming, but similar.
I have a theory that prediction markets will becomes hyperstition engines. This is a rather... violent example.
Do not enforce invariants with an LLM. Do not enforce invariants with an LLM. Do not enforce invariants with an LLM. Do not enforce invariants with an LLM.
Entire product or a feature for a product? Sometimes you just want to test an idea and vibe coding works well for that in the very short amount of time it takes now. Product market fit, user testing, engineering, those can come after the hunch.
Wait until you meet a VP in finance.
Oof. I turned the history referencing off. I use ChatGPT for wildly diverging topics and it will bring things up that have zero relevance to what I'm currently looking for if history is on.
This is a good idea. Not sure what ToS it would violate. But a good idea.
I think the harnesses are responsible for a lot of recent gains.
To pedant it up, not across GPUs.
Tulsa used to have a rich Black neighborhood.
SAP or Salesforce can do anything. Or you can write custom code for a fraction of the price for small/medium sized businesses.
If he's having an impact it's because he can break through the bureaucracy. He's not trying to protect a fiefdom.
LLMs fail at causal accuracy. It's a fundamental problem with how they work.
I can't do that either without opening up an image editing tool. Give the model a tool and goal with "vision". Should work better.
The reward is helping people hone their craft and grow in their careers. Being a shit shield sucks, but the trade-off is worth it.
Some people write to convince. Some people write to provoke.