MichaelDickens

4 ms·

1.
by MichaelDickens 8d ago
It has access to its own instructions, right? Otherwise how would the instructions work?
2.
by MichaelDickens 8d ago
> But there is one huge question: did Tristan opt out of model training for his ChatGPT and Codex sessions? If the answer is no, then this seems fair game. Just because something is legal and permitted by terms of service doesn't me
3.
by MichaelDickens 16d ago
I don't work on open source but it could be: - Good design requires user interviews, which are less common in open source - The sorts of people who are most attracted to open source work are programmers, not designers - Open source sof
4.
by MichaelDickens 17d ago
Zvi (author of OP) wrote several earlier pieces about the human failures at OpenAI. The most concise one is here: https://thezvi.wordpress.com/2026/08/08/what-happened-openai...
5.
by MichaelDickens 19d ago
What's wrong with Fandom?
6.
by MichaelDickens 29d ago
Well, I'm an American, but US media is disproportionately set in Los Angeles or NYC, and I haven't been to either of those places. I was watching Monk recently, and it feels kinda neat seeing a show shot in San Francisco because I
7.
by MichaelDickens 1mo ago
> Or they won't go to the trouble of using any browser that isn't the one provided by default by the OS? Chrome isn't the default on Windows or Mac, but it's far more popular than Edge or Safari.
8.
by MichaelDickens 1mo ago
Pangram says "100% AI-generated". I trust Pangram more than I trust this author's word.
9.
by MichaelDickens 1mo ago
Three years ago, people were saying "LLMs just generate plausible-sounding text, they don't understand the notion of truth so they can't do verifiable work like math proofs."
10.
by MichaelDickens 2mo ago
> I don’t know that the lack of information should lead you to assume that the Mexico lawsuits are justified. Gricean implicature.[1] The article wants to make the food companies look like the bad guys. If the lawsuits were clearly unjus
11.
by MichaelDickens 2mo ago
Following rz2k's suggestion, you could have a building employee operate the elevator during busy times.
12.
by MichaelDickens 2mo ago
Yes, this is an argument that exists. But it's not supported by evidence. It's the same as the old "video game violence should be outlawed because it might cause real violence", which is just as unsubstantiated.
13.
by MichaelDickens 2mo ago
YC founders aren't the ones taking risks. The risk comes from pouring your life savings into a company and not getting into any incubators and never getting any VC funding and then having no savings and nothing to show for it.
14.
by MichaelDickens 2mo ago
I would rather see you engage with the substance of the article rather than skipping right to insulting the authors. I don't think this sort of comment is up to the standard I have come to expect from HN.
15.
by MichaelDickens 2mo ago
> OpenAI apparently used confidential Apple hardware information when approaching Apple suppliers, and tricked one company into using a "specific trade secret metal-finishing technique" for an OpenAI device by claiming it had A
16.
by MichaelDickens 2mo ago
I read the main section of the lit review linked by OP, and it didn't seem to come to any real conclusions. > The results of this review have to be interpreted with caution because the number of studies in each category of cough pre
17.
by MichaelDickens 2mo ago
> It's not even available behind the counter anymore, presumably because they can make more money from morons buying the placebos. What do you mean "morons"? Say I'm a normal person who doesn't habitually read ma
18.
by MichaelDickens 3mo ago
> An honest question: Would you like to live in a world where company/employer exerts more control over the views that are publicly expressed? I don't really object to you asking this question, but I do object to you calling a
19.
by MichaelDickens 3mo ago
If that's true, then it suggests an easy fix: leave your application up for four hours, then discard all applications you get for the first two.
20.
by MichaelDickens 3mo ago
OpenAI and Anthropic valuations are based on the premise that they may develop AGI in the near future. How do you value a company based on that premise? Throwing regulations into the mix doesn't make the problem much harder than it alr
21.
by MichaelDickens 3mo ago
The trolley problem is probably the most famous thought experiment in ethics, but it doesn't do a good job of distinguishing ethical theories. Both utilitarianism and Kantian deontology agree that it's correct (or at least permiss
22.
by MichaelDickens 3mo ago
They aren't doing anything like that. In fact, they used to specifically train LLMs not to say they're conscious, because users didn't like it. (Maybe they still do that, all I know is they used to.) AI companies' ince
23.
by MichaelDickens 3mo ago
Doesn't seem pedantic to me. It's responding to the central thesis of the parent comment.
24.
by MichaelDickens 3mo ago
I distinctly remember buying my first SSD laptop in 2012 and being amazed at the boot time. At the time, I didn't realize it was because of the SSD, I just thought, "wow, the devs have made some fantastic optimizations in boot tim
25.
by MichaelDickens 3mo ago
Unless I'm missing something, this doesn't seem hard to fix: just let users decide whether hidden numbers should be ignored or received.
26.
by MichaelDickens 3mo ago
What an odd insult to identify a group of people solely by the fact that they read (a particular subset of) books, and then say they can't read. The ability to read is one of the only things this group of people has in common!
27.
by MichaelDickens 3mo ago
FWIW I understood your point just fine. It seemed to me that you made a clear enough distinction between "evidence that Claude didn't increase bugs" and "no good evidence either way".
28.
by MichaelDickens 3mo ago
What do you mean "better profitability criteria"? I don't want an index to exclude companies on the basis of profitability. I want it to hold the market. (I also don't want them to create special exceptions. The S&P
29.
by MichaelDickens 4mo ago
There is an entire segment of mutual funds and ETFs that avoid stocks with high valuations, called value funds. About half my equities are in value funds. The downside is that it's not an arbitrage. Sometimes they perform worse than th
30.
by MichaelDickens 4mo ago
I wouldn't say Anthropic is worse than OpenAI, but there's a lot wrong with them. https://anthropic.ml/ has a collection of incidents and relevant evidence.

More ›