esafak

3 ms·

61.
by esafak 12d ago
Embed a constitution they can't override. Project bad outputs to their nearest acceptable one. If we have to stop model development to ensure we can do it, so be it.
62.
by esafak 12d ago
Because all the models train on synthetic data too, plus they can collect their own data with robots.
63.
by esafak 12d ago
Yours put your laundry and dishes away?
64.
by esafak 12d ago
Housing and healthcare are going to get cheaper??
65.
by esafak 12d ago
Alignment is the architectural solution. Make it so the model can't misbehave. Yet many people here deride it as tainting the model, claiming "Whose values is it aligned to?" Sandboxes are a last ditch layer. They fail, as
66.
by esafak 12d ago
The cost of the code should be weighed against its benefits; at design review, code review, and launch review.
67.
by esafak 13d ago
Yes, it was: > What should the AI do when asked if it should nuke a country?
68.
by esafak 13d ago
And kids don't need parents to eat their vegetables and do their homework.
69.
by esafak 13d ago
Run and present the numbers, then defer. This is not rocket science.
70.
by esafak 13d ago
YC strongholds like Knox County: https://www.wvlt.tv/2026/09/01/knox-county-commission-passes...
71.
by esafak 13d ago
You are not making an argument for open models, but aligned models. There is no reason an open source model should be aligned. In fact, people often prefer open models because are not aligned, which they view as undesirable fiddling.
72.
by esafak 13d ago
I suggest putting the tests in a separate folder ;)
73.
by esafak 13d ago
Safety of humans!!! Simple things like not getting killed or enslaved. We could start there...
74.
by esafak 13d ago
Even the Chinese ones, which have no IPO gymnastics?
75.
by esafak 13d ago
They imitate humans. Alignment is about shaping their behavior towards safety.
76.
by esafak 13d ago
wtf?? P(doom) relates to https://en.wikipedia.org/wiki/Existential_risk_from_artifici... How would open source models bring about a MAD between humans and AI? By enabling us to wield aligned AI against nonaligned AI? I
77.
by esafak 13d ago
The idea is that today's companies don't go up after the IPO; only investors benefit from it.
78.
by esafak 14d ago
Probabilistic safety is too low a bar. We need to project the model outputs onto a safe subspace; lobotomize them, if you will. It may make them dumber, but that's a fine price to pay. I don't work in this space so I don't kn
79.
by esafak 14d ago
How long do you think that's going to last? A few more years, at most. Wail 'til the Chinese ramp up production.
80.
by esafak 14d ago
> And which experts will the government ask? Oh, right, the parties that stand to benefit the most from regulation: OpenAI and Anthropic! You've never heard of academia, NGOs, and intergovernmental organizations? Sure, the AI labs w
81.
by esafak 14d ago
Why is it so hard for people to understand the concept of agency in 2026?? What did your CPU do without instruction? Did OpenAI tell its agents to hack the orgs it did? Come on, make a half sensible argument.
82.
by esafak 14d ago
This makes no sense. The hardware is only going to get cheaper. In a decade, everybody could be running an ASI on their phones. What we need to do is to make unaligned AIs illegal and monitor for them, like we do with nuclear weapons. And m
83.
by esafak 14d ago
I was in Amish country last week, watching in bemusement as elderly women zoomed down the street in e-bikes. I thought to myself that it was understandable given that they lived on uneven terrain. But then, I wondered, why did they choose s
84.
by esafak 14d ago
> We should never have built machines with agency... We haven't quite crossed that bridge, but we do appear to be standing on it.
85.
by esafak 15d ago
What's interesting is that these are the smartest people in the world, and AI is eating their lunch. You do the math -- pun intended.
86.
by esafak 15d ago
Which mathematician's take on AI do you like? Do you think mathematics should no longer be an intellectual pursuit?
87.
by esafak 15d ago
This person is responding sensibly. I wish him well. What remains to be done is to accept that change has come, and take part in shaping the future. As for programming I am getting more done than ever before; I just outsource the implementa
88.
by esafak 15d ago
Is there a Github issue to track it? Also, the landing page footer is hard to reach due to infiniscroll, plus the Help overlay obscures two of the links.
89.
by esafak 15d ago
And it can -- if an able person wields it.
90.
by esafak 15d ago
Aka https://mobbin.com/ , which now has https://mobbin.com/mcp

More ›