damowangcy

I don't have the budget to do so, but if I do, it would be interesting to put different models in the same game with the same prompt to observe the behaviors that different models have when told to problem solve.

"AI found decompiling game to fix a bug after being instrcuted to beat it" sounds like an IPO friendly headline.

Fable seems to be a potential franchise to do all this; imagine a "living fantasy world" where everything truly unfolds based on your decisions, giving NPCs all the functions/actions that a player has.

In a dystopian fufure where a few or ONE AI company bested the rest, the company might banned you from using the AI for coding because the AI thinks you're incompetent or require you to pass a test to activate the coding function, because it's dangerous.

Assuming that there's still SWE jobs, when you attend an interview, the AI interviewer from the same AI company may just reject you because you asked which IDE to use just a week ago and it thinks you're not fit for the job.

I think this is what the whole "AI killed software" fallacy came from.

>nearly the same amount of effort to post a complaint you can delegate the task of reproducing the issue, diagnosing the root cause, and proposing a fix.

Is the effort the same though?

And even it is the same, making a meaningful contribution to an open source project doesn't mean just fixing your problem, but you also need to stay in line with the contribution guidelines, the project's direction and many more, which takes much more effort than just vibe coding a version for yourself.

Building software for one and for a bunch are two different things. AI is good at the former and can be used for the later but not without much effort. Models will get better, true, but that doesn't mean we need less engineering, in fact we need more to stay up-to-pace with AI. Imagine everyone submitting a PR to fix a bug, how fast can you review it manually? And if you use AI to review/verify them, the cost just shift from person who submitted the PR to the open source maintainer.

That doesn't mean AI is bad but that we need new systems and rules.

At this point of time, I just want a display.

Wait, you mean I can run a local open source LLM using me and my neighborhood idling smart IoTs?

That sounds like a good business model.

Marketing.

Have you guys see how aggressive is the push for enterprise use by both OpenAI and Anthropic? I had friend from a non-tech industry in Asia telling me that their company was offered free trial of the enterprise version of Claude, with trainings and such.

On the other hand, DS and Z.ai, have zero to none marketing outside China. There is friction to use DS/GlM models and the ZDR is unclear, so most enterprise that has heavy AI usage hasn't move over yet. They would rather spent $200 for the peace of mind than to take the risk of being slam as a national traitor down the road (which again is another form of marketing by Big AI, trying to frame Chinese models as thiefs).

So, I don't think they are not freaking out, it's just that they are addressing different market segments and reacting to the situation differently.

"Meet the F1 developer who fix a critical bug in 50 minutes with AI agent"

2000 PRs per day btw.

remind me in a few years when the same CEO is writing long post on Linkedin explaining why they need to let go half of the company.

tbh, they don't really need the website, it could just be a simple pure text website with a simple tagline explaining what they do, a sign in and sign up button, that's it.

what really matters the most is the dashboard.

We should be worry about both but at this point of time, agents aren't independent intelligences, someone has to use/create it.

A bad analogy but if your dog bites, we don't frame the dog as having awaken some unexplainable ability to defy your orders (your dog went rogue). Just because you spent lot of time to train it to not bite upon your instruction, doesn't mean it's the right training protocol or that you introduce enough variable or gave the right instruction. You're lawfully responsible for the bite and ethically wrong for removing the leash on your dog to test whether it bites(do it enough time, you now have a criminal intent), worse you didn't even closely monitor it.

Well, they can still farm, like the rest of us.

We engineered the conditions. The agents just obey. It didn't go rogue. The specs were wrong.