SalaciousBreadCrumb

[SalaciousBreadCrumb @BreadSalacious]

@FitzFromRWpod @WeatherProf The guy never claimed it was the largest anomaly in the last 300 - 500 years. He literally said he has never seen an anomaly so large "in his career". Do you think his career has lasted 300+ years, or are you just straw-manning?

[SalaciousBreadCrumb @BreadSalacious]

@tunguz The Big Lebowski. Also, just as good on a sunny Monday morning.

[SalaciousBreadCrumb @BreadSalacious]

@avappleyard Used his speed to cook the first defender. Then took it to the right side of the ice so he could isolate Jack Hughes (a forward) 1 on 1. Walked him easily. Only a perfectly-timed poke-check by Hellebuck prevented a Grade A chance. Not a selfish play by McDavid. A smart one.

[SalaciousBreadCrumb @BreadSalacious]

@Watashi_Yemox Q: What is an easy way to tell if someone is a racist?

A: They are upset by the skin color of an actor in a television show or movie.

It is a fucking TV show. An actor's skin color has zero impact on the quality of your life. The source material is irrelevant.

[SalaciousBreadCrumb @BreadSalacious]

@ibab @joannejang Wow, the stars really aligned for that excuse. First, unprompted, an altruistic employee decided to defend Musk and Trump's honor. Second, the code reviewer somehow missed this important change to the system prompt. Culprit being ex-OpenAI is just the icing on the cake. Right...

[SalaciousBreadCrumb @BreadSalacious]

@elder_plinius Sure...🙄

[SalaciousBreadCrumb @BreadSalacious]

@realDonaldTrump Donald J. Trump likes his chicken spicy.

[SalaciousBreadCrumb @BreadSalacious]

@GaryMarcus @kevinroose You're happy to quote Chollet when he says this is not AGI, but keep dismissing that he also said, "The fact that the model was ARC-adapted via the training set absolutely does not invalidate its score." O3 performed the test within the bounds of his ruleset.

[SalaciousBreadCrumb @BreadSalacious]

@GaryMarcus @hal_incandenza4 This obstinance has reached an absurd level. At this point your contrarian take smacks of engagement farming rather than any real conviction. True to form, I suppose. I'm sure I'm not the only one who remembers your leaked DM to Strawberry Man.

[SalaciousBreadCrumb @BreadSalacious]

@GaryMarcus All this poll will show is that anyone who selects 'No' also doesn't understand the SOP for testing on this benchmark. Just take the L and admit you were wrong, like a grown-up, instead of doubling down on your erroneous take.
https://t.co/73CwGBRTJy

[François Chollet @fchollet]

@NielsRogge This is the point of the training set, though -- to train your model on it. It would be even more impressive if the model had no prior exposure to ARC data, but the fact that the model was ARC-adapted via the training set absolutely does not invalidate its score.

[SalaciousBreadCrumb @BreadSalacious]

@OfficialLoganK Would love it if while using it to code a game it could also generate the graphical assets (sprites, backgrounds etc) for the game, then allow the user to download the graphics along with the code, put them in the appropriate directory, and have it run the game seamlessly.

[SalaciousBreadCrumb @BreadSalacious]

@elonmusk This from the so-called "Free Speech Absolutist". The hypocrisy is breathtaking.

[SalaciousBreadCrumb @BreadSalacious]

@RMPutnam25 @lmichael_98 @RevelAidan @SnazzyLabs @elonmusk They are clearly not doing that, so why even bring it up? Why getting everyone riled up by tweeting out his hypothetical response to another company's hypothetical actions?

[SalaciousBreadCrumb @BreadSalacious]

@GaryMarcus OpenAI had a massive lead and others have only now managed to catch up. There haven't been any training runs significantly bigger than GPT-4 yet. Why is that so hard for you to comprehend? Only If GPT-5 (or Claude 4) isn't measurably better than GPT-4 is your point valid.

[SalaciousBreadCrumb @BreadSalacious]

@PYRAMIDB0Y @3_deame @teslatidbits @bcmerchant @danahull Claiming that I don't know how to raise a child based on a single paragraph is the absolute height of arrogance. There are thousands of ways to make a positive connection with one's child. Giving them a super cool card, designed by AI (oh, the horror!) could be one of them.

[SalaciousBreadCrumb @BreadSalacious]

@PYRAMIDB0Y @3_deame @teslatidbits @bcmerchant @danahull Have you never met a child? If I give my kid a cupcake he is stoked because he loves sugar. He doesn't care if I baked it for him or I "half-assed it" and bought it at the grocery store. That kid won't care a whit about a 0 effort gift. He'll be happy because he got a cool card.

[SalaciousBreadCrumb @BreadSalacious]

@GaryMarcus @OpenAI Did you read the part about them (and Musk) realizing they were going to need much more capital to achieve their goal? Smart leaders pivot when unforeseen realities become apparent. Do you expect them to drive the company into the ground, stubbornly staying true to the mission?

[SalaciousBreadCrumb @BreadSalacious]

@per_anders @DrJimFan 🙄Shame on Gutenberg for putting the scribes out of work. Shame on Edison for putting the candlemakers out of work. Shame on Ford for putting the carriage drivers out of work. Technology marches on, as it ever has, to the great benefit of mankind. Don't whine about it. Adapt!

[SalaciousBreadCrumb @BreadSalacious]

@joshyoung Wow, that is a dumb take. What's relevant is the underlying dryness of the forests. They're tinderboxes waiting to explode by *any* means of ignition. Or do you suppose arsonists are going into the woods with hairdryers and drying out thousands of hectares before tossing a match?

[SalaciousBreadCrumb @BreadSalacious]

@FortuneForge That is an incredibly poor analogy. Chess is played within a set of immutable rules, against a single opponent also bound by those rules. Neither player starts with any advantage. There are no elements of chaos or randomness at all.

Life is the opposite of all of that.