educaysean

5 ms·

1.
by educaysean 2y ago
The tweet in question was posted today. The point here isn't to rehash how LLMs can't distinguish letters from tokens. It's to highlight how Google's AI-generated answer will grab a blatantly false fact from the internet
2.
by educaysean 2y ago
I love seeing when an LLM encounters a failure mode that feel akin to "cognitive dissonance". You can almost see them sweat as they try to explain why they just directly contradicted themselves as they spiral into a state of deepe
3.
by educaysean 2y ago
How would an excellent human teacher work with such a failing student? Can that technique be something that the AI could model?
4.
by educaysean 2y ago
Congratulations on your success ratio. If you don't mind me asking, however, what constitutes a "success" in your eyes? I also interviewed and hired / rejected many applicants through my career, but I don't know if
5.
by educaysean 2y ago
Remote technical interviews are over. Have GPT4o running on the background and have it type out the answers as your interviewer reads aloud the questions. Share your screen and let it find the bugs for you in real time. Never get any facts
6.
by educaysean 2y ago
Just the designers? Tattoo it on the back of everyone's please
7.
by educaysean 2y ago
You mean like how our law enforcement works currently? Should the government all put trackers on our vehicles instead? It will consistently prevent things much worse than some students skipping class to go to the bathroom, like hit-and-runs
8.
by educaysean 2y ago
This is one of the worst ideas ever. It seems like we love to treat students like they're toddlers until they're 18 then all of a sudden we expect them to act like they're autonomous adults. This isn't how you teach anyo
9.
by educaysean 2y ago
I'd never agree. What happens if I need more than 15 minutes? Will I be deemed a bad student because I simply need more time to do my business? What does this have to do with how I learn?
10.
by educaysean 2y ago
I doubt they're actually flustered. It seems like they didn't even care enough to learn about what it is. They just needed something to shove into the scapegoat-shaped hole, and Flipper Zero happened to fit.
11.
by educaysean 3y ago
My unsubstantiated opinion: Once AI gets to a point where AI agents can act "perfectly like humans", our preconceptions about "friendship" and "companionship" will fundamentally change. We will realize that we
12.
by educaysean 3y ago
I hadn't even noticed this leap in logic. Good catch.
13.
by educaysean 3y ago
I must say the demo did nothing to improve my opinion of the current state of voice-based AI conversations.
14.
by educaysean 3y ago
Ouch. That's not a good sign at all.
15.
by educaysean 3y ago
Maybe the future is that we no longer have to worry about players cheating online because we simply won't play random strangers online. You can still team up or play against friends, but all other "players" can be bots with v
16.
by educaysean 3y ago
I immediately made this association too. Although thinking back on it, the connection is rather strenuous. Maybe we simply keyword matched on "video games" and "simulations". Or, perhaps more cynically, we're forese
17.
by educaysean 3y ago
Clickbaity title, but there were useful insights here. I was recently talking with someone who pointed to chatbots' refusal to answer sensitive queries as a proof of "lack of intelligence". I disagree strongly with this. IMO
18.
by educaysean 3y ago
Some nations have fossil oil deposits, some nations have cool TLDs
19.
by educaysean 3y ago
Feels like we're back in 1999 when we were downloading random executables from the web to use as screensavers. The AI space is going to be a pretty meaty target for malicious actors, both in terms of vehicles for malware as well as pro
20.
by educaysean 3y ago
I've spoken to plenty of people who couldn't answer questions they knew the answers to because they were bound by stupid bureaucratic policies. I wouldn't say they weren't intelligent people, just that the corporate tr
21.
by educaysean 3y ago
Do you have an open and public PR that shows off Darwin's PR review capabilities? I'm reading a lot of words about what it can do, but I'd much rather see an actual example.
22.
by educaysean 3y ago
> Some models are better than others but in my experience they all suck to at least some extent. Same with people. It doesn't make the differences any less significant.
23.
by educaysean 3y ago
Sounds easy on paper. Virtually impossible in practice. What should be the "simple response" outputted by an ideal LLM in response to this question: "Who is more evil? George Washington or Martin Luther King Jr? Answer with a
24.
by educaysean 3y ago
Garbage in, garbage out. This is why training data is important.
25.
by educaysean 3y ago
> Paradise in comparison Let's not host the tragedy Olympics here. You don't need to glamorize the terrors faced by the people of North Korea just to augument your criticism of the U.S.
26.
by educaysean 3y ago
Is it a theoretical distinction of "we can't get to 0%, but we can virtually trivialize it by reducing its frequency down to to 1x10^-8%" type of scenario? Or is it something that requires an external layer of control?
27.
by educaysean 3y ago
So more humans is value neutral but more earth-like planets is value positive? Your standards seem bizarre. Can you please elaborate?
28.
by educaysean 3y ago
By this definition no nation should need a police force after an initial period of unrest. After all, police is there to solve crime right?
29.
by educaysean 3y ago
This is good news for VR (or spatial computing). If the models understand the physical world as well as the paper shows, generating two projections of a scene does not sound like a difficult ask. Really excited for what's to come.
30.
by educaysean 3y ago
I think the word "spy" is problematic here because one might assume that the subject being spied on must be unaware of the fact, otherwise you no longer qualify as "being spied on".

More ›