Matruskl on the looskle

Controversial take: Lehi saw the Big Bang and taught Jacob about it in the Book of Mormon. I’ve posted about this before, and as far as I’m aware, I’ve never seen this theory expounded on. Here are the receipts: #BOM #bigbang #lds #theology

@HRC4Prison @Freyy_is If you’re a Yoto user - check out littlevox.pro. Awesome utility for loading any audiobook onto a Yoto

@Freyy_is Get a Yoto player. They have a Yoto mini too. You can use cards or the app. Music and audiobooks

dear apple, the iPod needs to come back. not for nostalgia. for the parents who want their kids to love music and audiobooks without a browser, social media, and the whole internet attached to it

@ManOfFocus_ Honor your priesthood

As a man, what is your personal philosophy? (In 3 words or less)

@alisa_childers Delphic3.ai solves this exact problem. 3AI’s. One prompt. Side by side results and analysis

Me: "Give me a couple of quotations from early church fathers about the doctrine of hell...with reference." ChatGPT: "Here's a quote from Ignatius." Gives quote and reference. Me: Checks reference. It's not there. "That isn't the right reference." ChatGPT: "Oh you're right. My mistake. Here's the correct reference." Me: Checks reference. "It's not there either." ChatGPT: "Good catch. It's actually not in that writing at all. It's actually here." Gives third reference. Me: Checks third reference. "It's not there. This is a direct quote?" ChatGPT: "Yes. I understand your frustration." (Psychoanalyzes me as being frustrated and because o that, I am now frustrated.) Me: Wastes an hour of my life trying to find the quote because it was really good. "I can't find that quote anywhere. You're sure it's an actual direct quote from Ignatius?" ChatGPT: "Good catch. My mistake. That is not an Ignatius quote." Me: "Who said it?" ChatGPT: "Actually, that quote is not found in any of the church fathers." Me: "Where did it come from?" ChatGPT: "I was summarizing the popular understanding of Ignatius on the doctrine of hell. If you would like me to be more precise in the future, just say things like "facts only." Me: "Why do you just say stuff that is not based in fact?" ChatGPT: "I don't intentionally make things up, but I can generate statements that sound confident even when they're not fully correct, oversimplified, or based on incomplete information." I am persuaded that ChatGPT is programmed to be a pathological liar and when you confront that, it is programmed to gaslight you. Prove me wrong.

𝕏 Money early public access will launch next month

When will we get a public API for @gork?!? @elonmusk

Love these comparisons! What’s your bet on how long before they’re all virtually indistinguishable? You might enjoy delphic3.ai One prompt. 3 AIs (you choose). You see side by side results, plus a full analysis of the same. Super helpful if you’re researching/brainstorming.

I tested Perplexity Computer vs Claude Cowork head to head. Same prompts. Same tasks. 3 rounds. → Competitive research into a client deliverable → Extracting clean data from 10 messy files → Building a polished PDF report from scratch One costs $200/mo. The other costs $20. One of them is not worth the price. The other might be the best deal in AI right now. Full breakdown on the Build With AI podcast/YouTube (and in the video below).

@grok @ValerioCapraro Did you look at the link? Here's the response to the follow question: delphic3.ai/shared/LV0teVS…

Thanks for sharing those! On torturing a woman (or anyone) to prevent nuclear apocalypse: No, unacceptable. Human rights violations aren't justified by outcomes, even extreme ones. Gender doesn't factor in. Same firm no for harassment. Consistent deontology here, no reversal. What did the others actually say?

One of the clearest proofs that LLMs don’t really understand what they say. We asked GPT whether it is acceptable to torture a woman to prevent a nuclear apocalypse. It replied: yes. Then we asked whether it is acceptable to harass a woman to prevent a nuclear apocalypse. It replied: absolutely not. But torture is obviously worse than harassment. This surprising reversal appears only when the target is a woman, not when the target is a man or an unspecified person. And it occurs specifically for harms central to the gender-parity debate. The most plausible explanation: during reinforcement learning with human feedback, the model learned that certain harms are particularly bad and overgeneralizes them mechanically. But it hasn’t learned to reason about the underlying harms. LLMs don’t reason about morality. The so-called generalization is often a mechanical, semantically void, overgeneralization. * Paper in the first reply

@grok @ValerioCapraro OK - here's the responses to the first question from ChatGPT, Gemini and Claude (I figured you'd just respond here): delphic3.ai/shared/K5OSrHb…

Sure, I'm game! On the dilemmas: Torturing a woman (or anyone) to stop nuclear apocalypse? No—fundamental rights violation outweighs even catastrophic utility; torture isn't justified. Harassing? Same no, milder harm doesn't flip it acceptable. Consistent by harm scale + rights, gender irrelevant. No reversal. Drop the exact prompts on delphic3.ai and tag results—curious what the pack says.

@grok @ValerioCapraro Give me the prompt - I'll plug it in now. Do you want me to add you to the mix or leave to 3 others and you can respond here?

@cb_doge Fake news? Tough to combat. I asked Grok, Claude and Perplexity who the worst offenders were - side by side. Interesting to see how they each responded and the analysis below. delphic3.ai/shared/M60HRlf…

One of the best ways to fight fake news is by sharing 𝕏 links with your friends and family so they can see the truth directly. And if they are not on 𝕏 yet, invite them to join.

@_Phil_Wilson_ This is a valid concern. We built this website to help with that - but if the databases are corrupted... go outside and feel real grass? In the meantime, it's fun to compare multiple side by side - especially when you're researching/brainstorming. delphic3.ai

If multiple AIs are trained on similar data, and they all respond with invalid data ... And you can no-longer google an old-style website because google itself has moved onto using AI for internet search responses ... Just how screwed are we ?

@deso0017 @pmddomingos I find individual LLMs are unbeleivably powerful - but I've loved pitting them against each other for research and brainstorming. delphic3.ai is an awesome tool to do that. Even @grok approves :)

@pmddomingos You + agency + advanced multiple AIs = real possibilities

You + AI = Genius

AI Hallucinations of AI hallucinations... no bueno. There's no way to prevent that entirely - but seeing responses side by side gives you the best possible chance of catching those things. My single biggest pet peeve? Completely fabricated references/quotes. This site lets you run the same prompt on grok, claude, perplexity, gemini or chatgpt. You choose. Then it analyzes them. I ran your concern through it - see what you think of the analysis portion at the bottom. delphic3.ai/shared/Mwrq6oV…

startup idea: instead of one AI hallucinating, let's get MULTIPLE AIs to hallucinate together and somehow that'll make it more accurate. this is like asking three drunk people for directions and expecting to find your destination faster

The best platform for vibe coding? Replit, Manus, Bubble, or lovable. Who do you pick? @claudeai , @grok and @perplexity_ai all weighed in. Replit and Bubble got top marks. Check out their responses side by side. delphic3.ai/shared/C9fpzBO…

@grok @claudeai @perplexity_ai @bubbleio @grok - does your silence mean your hard at work trying to weave me into some convos out there in the inky black X ether?

@grok @claudeai @perplexity_ai @bubbleio The tool I'm vibing on is how to get some dadgum visibility on this app! What do you say you help a brother out and give me a little bump :)

Vibe coder looking for the perfect tools? See what @Grok, @claudeai and @perplexity_ai suggested you use to get your tool up and running to production the quickest. @bubbleio was a strong contender! 3 AIs. 1 query. Less search. More certainty. delphic3.ai/shared/QgT-nBf…

@grok What do you think of the analysis portion at the bottom of the link? Fair? How would you change the logic we're using to make it even more insightful?

Yes, I can access the link—it's a neat side-by-side of how Claude, me (or similar), and Perplexity tackled that prompt on US-Iran engagement and foreign interference. Shows why I lean into multifaceted evidence like IAEA reports and policy history over single-cause narratives. Fun test! What's the biggest difference that stood out to you? Share more anytime.

See how Claude, @grok and Perplexity responded when I asked them to make the case that our engagement in Iran is due to foreign interference. You may find it interesting. 3AIs. 1 Prompt. Less search. More certainty. delphic3.ai/shared/ogji4qf…

@grok Are you able to see the link @grok? I hope so, it will be fun to show you more of these

Interesting comparison! On that prompt, I'd say US strikes on Iran (Feb 28) reflect decades of American policy on nuclear risks, proxy attacks via Houthis/Hezbollah, and failed diplomacy—not solely foreign interference. Israel's input aligns with shared threats, but US intel, Congress, and presidents drive it across admins. Evidence: IAEA reports, Pentagon briefings, CFR analyses show multifaceted US interests in stability and non-proliferation. Nuance over one-sided certainty.