I have no clock in my apartment. To find out what time it is, I need to fire up one of my computers and look. Sometimes it’s the kindle
- 2 Posts
- 3.66K Comments
intensely_human@lemm.eeto Technology@lemmy.world•Apple just proved AI "reasoning" models like Claude, DeepSeek-R1, and o3-mini don't actually reason at all. They just memorize patterns really well.English32·27 天前They aren’t bullshitting because the training data is based on reality. Reality bleeds through the training data into the model. The model is a reflection of reality.
intensely_human@lemm.eeto Technology@lemmy.world•Apple just proved AI "reasoning" models like Claude, DeepSeek-R1, and o3-mini don't actually reason at all. They just memorize patterns really well.English21·27 天前Computers are better at logic than brains are. We emulate logic; they do it natively.
It just so happens there’s no logical algorithm for “reasoning” a problem through.
intensely_human@lemm.eeto Technology@lemmy.world•Apple just proved AI "reasoning" models like Claude, DeepSeek-R1, and o3-mini don't actually reason at all. They just memorize patterns really well.English21·27 天前I appreciate your telling the truth. No downvotes from me. See you at the loony bin, amigo.
intensely_human@lemm.eeto Technology@lemmy.world•Apple just proved AI "reasoning" models like Claude, DeepSeek-R1, and o3-mini don't actually reason at all. They just memorize patterns really well.English1214·27 天前Fair, but the same is true of me. I don’t actually “reason”; I just have a set of algorithms memorized by which I propose a pattern that seems like it might match the situation, then a different pattern by which I break the situation down into smaller components and then apply patterns to those components. I keep the process up for a while. If I find a “nasty logic error” pattern match at some point in the process, I “know” I’ve found a “flaw in the argument” or “bug in the design”.
But there’s no from-first-principles method by which I developed all these patterns; it’s just things that have survived the test of time when other patterns have failed me.
I don’t think people are underestimating the power of LLMs to think; I just think people are overestimating the power of humans to do anything other than language prediction and sensory pattern prediction.
intensely_human@lemm.eeto Technology@lemmy.world•Amazon is reportedly training humanoid robots to deliver packagesEnglish51·30 天前humanoid robot: dances
amazon: shock
humanoid robot: makes coffee
amazon: shock
humanoid robot: delivers package
amazon: friendly shock
“eyesed” in the eyes of the literal
intensely_human@lemm.eeto Not The Onion@lemmy.world•Woman divorces husband after ChatGPT reads his coffee grounds and predicts affairEnglish31·2 个月前Shadowrun is old now, man
intensely_human@lemm.eeto Not The Onion@lemmy.world•Woman divorces husband after ChatGPT reads his coffee grounds and predicts affairEnglish51·2 个月前That’s a great question!
intensely_human@lemm.eeto Not The Onion@lemmy.world•Woman divorces husband after ChatGPT reads his coffee grounds and predicts affairEnglish102·2 个月前Easy for a man to get custody of the kids? What world do you live in?
My friend’s ex repeatedly crashed her car while drunk with their daughter in the car and he had to fight a pretty much non-stop court battle for over a decade to maintain his 50%.
intensely_human@lemm.eeto Technology@lemmy.world•Some Reddit users just love to disagree, new AI-powered troll-spotting algorithm findsEnglish1·2 个月前(they’re both wrong)
intensely_human@lemm.eeto Technology@lemmy.world•Some Reddit users just love to disagree, new AI-powered troll-spotting algorithm findsEnglish1·2 个月前Is it though? According to the book The Five Dysfunctions of a Team, conflict is a sign of a trusting team, and conflict is a necessity for that team to achieve commitment.
I can attempt to explain how it it’s unclear
This graphic is from the book:
intensely_human@lemm.eeto Technology@lemmy.world•Some Reddit users just love to disagree, new AI-powered troll-spotting algorithm findsEnglish11·2 个月前Dude your mom feels like bait
intensely_human@lemm.eeto Technology@lemmy.world•Some Reddit users just love to disagree, new AI-powered troll-spotting algorithm findsEnglish112·2 个月前Trolling and disagreeing are not the same thing.
I am someone who “loves to disagree”, as in most of my engagement on lemmy is disagreeing. But that’s because I speak up when I feel like speaking up is needed, and when I feel like it’s needed is when what I believe is not being said.
So this makes me wonder if one could force a move into a higher dimension by somehow constraining a set of connected distances in this way.
Sort of like protein folding as a way to bootstrap a dimensional jump.
intensely_human@lemm.eeto Fediverse@lemmy.world•Stop Internet Searching and Start Asking on Fediverse?English31·2 个月前Another idea is an asking system that compiles answers from the web while also posting the question here. Gives the benefit of immediate answers while also populating the fediverse.
intensely_human@lemm.eeto Mildly Infuriating@lemmy.world•Spotify 30 minutes of uninterrupted... Just kiddingEnglish41·2 个月前This episode is brought to you by Me Undies
intensely_human@lemm.eeto Ask Lemmy@lemmy.world•Is there homelessness in China? Why or why not? What happens in China if someone can't pay rent?0·2 个月前As a libertarian I appreciate their government’s not razing all that shit, and just letting people do what they need to do.
A friend of mine is homeless in SF. His mental health has dropped significantly ever since the city forcibly moved him from the spot where he was camping to a shelter. He says he was doing much better in the tent than in the shelter.
intensely_human@lemm.eeto Not The Onion@lemmy.world•Scrabble players quit game after 400 offensive words banned from listEnglish7·4 个月前It’s in urban dictionary. It means “to be honest”
I’m considering getting a clock for my kitchen