Any response from @cursor_ai regarding sudden crashes with no logs or trace on Mac? Happened for the 4th time in 6 weeks today. There are multiple threads out there on it across github and reddit. Any idea at all what's happening or am I being forced to switch?

we're living in the future of dev except everything is expensive and breaks. in the past 8 hours I had a completely silent crash of @cursor_ai (where i pay $200/mo) then a few hours later I got blocked by @AnthropicAI claude code (also $200/mo) and now @github is broken

My initial excitement for @OpenAI Codex is now gone. It completely fails to do anything moderately complex, ignoring requirements and taking huge shortcuts. Identical prompt into @cursor_ai Agent w/ @AnthropicAI Claude 4 Opus knocked it out of the park on the first shot.

Fact. It’s taken things to an entirely new level and my entire process, from coding to testing to deploying to debugging, is now changed forever.

Today, we’re announcing our $50 million Series B and that Admiral John M. Richardson (USN, Ret.), former Chief of Naval Operations, is joining our Board of Directors. With over $70 million raised to date, we’re accelerating toward full-scale nuclear battery demonstrations in 2026, enabling deliveries to maritime and space customers in 2027. Read more via @demarest_colin in @axios: axios.com/2025/05/14/zeno-po…

Using o1-preview from @OpenAI reminds of using the web on slow dial-up in the 90s

One pair of founders I talked to today were students at a top US university. I asked what percent of students don't use AI to write their papers for them. They said max 20%. Professors have given up trying to forbid it.

THIS IS #HOMEOFBEISBOL 🩵 See you at 4:10pm: marlins.com/tix

It is Saturday before Christmas and there are still AI products shipping.

Have been playing with Gemini Pro for our core use cases related to music. Unfortunately if there’s even a single curse word in the lyrics, it denies the request with no detail. Shockingly way more strict than GPT3.5/4 and of course OSS models.

This feels like someone has a time machine and identified the exact point in history to fork the timeline

I have been writing code in shell text editors for almost 30 years, mostly resisting IDEs (hey, java 🤷‍♂️). I'm not sure I will ever code again without both an AI typeahead and AI chatbot assistant. This shit is so wild and constantly blows my mind.

In 2 days there are have been 4 studies that help explain LK-99's potential superconducting abilities. These simulations converge on key properties that suggest a new class of SC materials, and help explain quirks of TK-99 we've seen so far. Here is the easy-to-digest summary so far: (Reminder: still no experimental replication yet!!) All studies converged on the fact that LK-99 has interesting electronic properties, formed by these 'flat energy bands' that can enable superconductivity through several different mechanisms. - This effect relies on copper replacing lead atoms in the crystal, but it has to replace very specific lead atoms for the bands to appear, meaning it may be hard to synthesize with high purity (paper 1) - The conduction pathways in the material may be one-dimensional, meaning they aren't equal in all directions, and this could be why it doesn't act as a perfect magnetic levitator but rather a semi-levitator. Also, other metals like gold could make LK-99 perform even better (paper 2) - TK-99 appears to be much more robust to disorder, or randomness in the crystal, while retaining its superconducting properties. And, it appears the overlap of copper and oxygen electron orbitals might explain why this occurs at ambient pressures (paper 3) - The most dramatic result of all is by the most distinguished author: the appearance of diamagnetism without superconductivity seems unlikely (paper 4). Here is the technical deep-dive to back it up: Note: I include highest h-indexes of authors, which is like a 'science high score' h=20: 'successful scientist', e.g. full professor h=40: 'outstanding scientist', e.g. fellowship in APS h=60: 'truly exceptional scientist', e.g. national academy 1st Paper - Lawrence Berkeley National Lab Sinead Griffith (h-index 20) arxiv.org/abs/2307.16892 @LBNLresearch scientist @sineatrix published the first simulation study on LK-99. Her study supported the original Korean team's (LKK) proposed mechanism whereby copper atoms replace lead atoms in a crystal structure, which introduces a twist or strain on the crystal. LKK measured a 0.5% volume contraction from this strain, which @sineatrix's simulation corroborated, but more interestingly this change in shape creates an interesting change in the available places electrons can fit into the material. These are known as 'energy bands' and normally are quite jagged, going up and down below the 'sea level' of energy like a mountain range. However, when these energy bands remain quite flat and stay close to sea level, aka 'fermi energy', it is thought to enable interesting properties like superconductivity, insulation, etc, and more. Notably these energy bands only form when one particular location receives the copper replacement atom, and it is the least-likely location. This suggests synthesizing the material may be difficult, with low-efficiency yield or purity. 2nd Paper - Shenyang National Laboratory arxiv.org/abs/2307.16040 Lai, Li, et al, and Xing-Qiu Chen (h-index: 47) These authors find similar results to @sineatrix, showing a volume contraction with lead-apatite has copper introduced, and which produces the signature 'flat energy bands' near the fermi level which are thought to make possible superconducting effects. They note that the available conduction pathways through the material seem one dimensional, meaning they do not extend in all directions equally. This might explain an imperfect Meissner effect where the material does not levitate perfectly, but only partially (my interpretation). Most interestingly, they try simulating similarly-sized atoms like silver and gold in replacement of copper atoms, and find that these maintain flat energy bands at the fermi surface. Other elements besides copper may improve the performance of LK-99-like materials. 3rd Paper - University of Colorado, Boulder arxiv.org/abs/2308.00698 Kurleto et al, Daniel S Dessau (h-index: 49) Also finds the same flat energy band structure at the fermi surface. More interesting, these authors simulate the effects of disorder being introduced to the crystal lattice, and see an interesting result: energy bands in TK-99 remain flat even when slightly disordered, i.e. not perfectly 'ideal crystal'. They suggest this could help explain how it retains superconductivity at such high temperatures. Also, they put forward an interpretation of the superconductivity as reliant on overlapping wave functions, or distributions, of the electron energy levels, i.e. the electrons are just slightly overlapping with each other, which creates the flat bands. They also claim the most important consideration is the overlap between the copper and oxygen electron orbitals, noting that the Oxygen-Copper pairing may explain why TK-99 may superconduct at much lower pressures than previous RTS materials (hydrides). (previous hydride-based RTS materials were superconducting at room temperature, but only at millions of atmospheres of pressure). 4th Paper - Northwest U and TU Wien arxiv.org/abs/2308.00676 Liang Si and Karsten Held (h-index: 67) These authors find the same results as the previous two studies - flat energy bands around the fermi surface as a result of copper atoms replacing lead. Two different methods for superconductivity are coupling between electrons, 'electron-electron', or coupling between electrons and vibrations in the crystal lattice, 'electron-phonon'. These authors find that both mechanisms could be possible given the structure that is formed. There is strong agreement with general findings of all the other papers mentioned, however they additionally claim the appearance of diamagnetism without superconductivity is at odds with their results. This would have immense implications for the early attempts at synthesis which show diamagnetism but have not yet yielded electrical measurements of zero resistance. ~~~~~~~~~~~ All the authors in these studies noted that their simulation results are not predictive or conclusive of TK-99 being the holy grail of materials science - a room temperature, ambient pressure superconductor. Taken together they paint an interesting and compelling picture - at the very least, TK-99 is a very interesting material that suggests new lines of research into superconductors that have the potential to perform at room temperatures and ambient pressures. Here is my mental model right now: This material is difficult to synthesize because getting the copper in the right place is not very likely, producing low-yields. However, it's possible to separate out individual flakes, or crystal grains, that are diamagnetic but these are quite small. Internal to these tiny crystal grains are one-dimensional 'superconducting' channels that result in partial flux-pinning / levitation, however because these don't enable currents to flow in any direction it isn't perfectly pinned. The energy bands that superconduct are resilient to slight disorder, and are oriented in such a way that lets them perform well without requiring large temperatures. The flakes that levitate aren't just diamagnets, but, the 1-dimensional conduction pathways might feasibly make it difficult to measure resistance (i.e. anisotropic resistance? - my speculation). ~~~~~ I'll reiterate that everything depends on experimental verification, but it is fascinating to read and share this rapidly developing scientific breakthrough. My personal odds keep going up.

At @vellum_ai, we've now worked with 50 customers in taking their LLM use cases to production. I'm excited to start sharing customer stories to bring our impact to life. Let's start with our friend @jgrayla at @ClapForEncore. Full story here: vellum.ai/blog/encore-increa…

Stable Diffusion SDXL 0.9 Available! The new stable diffusion model (SDXL 0.9) is now available on Huggingface hub. Thus model is an insane improvement in comparison to the previous versions. - The model: huggingface.co/nichijoufan77… - Refiner model: huggingface.co/nichijoufan77… - Colab notebook: colab.research.google.com/gi… --------- Run locally: - 𝚐𝚒𝚝 𝚌𝚕𝚘𝚗𝚎 𝚑𝚝𝚝𝚙𝚜://𝚐𝚒𝚝𝚑𝚞𝚋.𝚌𝚘𝚖/𝙵𝚞𝚛𝚔𝚊𝚗𝙶𝚘𝚣𝚞𝚔𝚊𝚛𝚊/𝚜𝚝𝚊𝚋𝚕𝚎-𝚍𝚒𝚏𝚏𝚞𝚜𝚒𝚘𝚗-𝚡𝚕-𝚍𝚎𝚖𝚘 - 𝚐𝚒𝚝 𝚕𝚏𝚜 𝚌𝚕𝚘𝚗𝚎 𝚑𝚝𝚝𝚙𝚜://𝚑𝚞𝚐𝚐𝚒𝚗𝚐𝚏𝚊𝚌𝚎.𝚌𝚘/𝚗𝚒𝚌𝚑𝚒𝚓𝚘𝚞𝚏𝚊𝚗𝟽𝟽𝟽/𝚜𝚝𝚊𝚋𝚕𝚎-𝚍𝚒𝚏𝚏𝚞𝚜𝚒𝚘𝚗-𝚡𝚕-𝚋𝚊𝚜𝚎-𝟶.𝟿 - 𝚐𝚒𝚝 𝚕𝚏𝚜 𝚌𝚕𝚘𝚗𝚎 𝚑𝚝𝚝𝚙𝚜://𝚑𝚞𝚐𝚐𝚒𝚗𝚐𝚏𝚊𝚌𝚎.𝚌𝚘/𝚗𝚒𝚌𝚑𝚒𝚓𝚘𝚞𝚏𝚊𝚗𝟽𝟽𝟽/𝚜𝚝𝚊𝚋𝚕𝚎-𝚍𝚒𝚏𝚏𝚞𝚜𝚒𝚘𝚗-𝚡𝚕-𝚛𝚎𝚏𝚒𝚗𝚎𝚛-𝟶.𝟿 - Option 1 - Low mem 𝙿𝚈𝚃𝙾𝚁𝙲𝙷_𝙲𝚄𝙳𝙰_𝙰𝙻𝙻𝙾𝙲_𝙲𝙾𝙽𝙵=𝚖𝚊𝚡_𝚜𝚙𝚕𝚒𝚝_𝚜𝚒𝚣𝚎_𝚖𝚋:𝟷𝟸𝟾 𝙰𝙲𝙲𝙴𝚂𝚂_𝚃𝙾𝙺𝙴𝙽={𝚊𝚌𝚌𝚎𝚜𝚜_𝚝𝚘𝚔𝚎𝚗} 𝙴𝙽𝙰𝙱𝙻𝙴_𝚁𝙴𝙵𝙸𝙽𝙴𝚁=𝚏𝚊𝚕𝚜𝚎 𝚙𝚢𝚝𝚑𝚘𝚗 𝚊𝚙𝚙.𝚙𝚢 - Option 2 - High mem: 𝙿𝚈𝚃𝙾𝚁𝙲𝙷_𝙲𝚄𝙳𝙰_𝙰𝙻𝙻𝙾𝙲_𝙲𝙾𝙽𝙵=𝚖𝚊𝚡_𝚜𝚙𝚕𝚒𝚝_𝚜𝚒𝚣𝚎_𝚖𝚋:𝟸𝟻𝟼 𝙰𝙲𝙲𝙴𝚂𝚂_𝚃𝙾𝙺𝙴𝙽={𝚊𝚌𝚌𝚎𝚜𝚜_𝚝𝚘𝚔𝚎𝚗} 𝙴𝙽𝙰𝙱𝙻𝙴_𝚁𝙴𝙵𝙸𝙽𝙴𝚁=𝚝𝚛𝚞𝚎 𝚙𝚢𝚝𝚑𝚘𝚗 𝚊𝚙𝚙.𝚙𝚢 ----- Images sources: -reddit.com/r/StableDiffusion… - reddit.com/r/StableDiffusion… - twitter.com/artificialguybr/…

Big Reveal 💥! Today, we're launching a Discord Bot that morphs your words into 3D models in under 2 minutes! Say goodbye to complex 3D software and hello to a new era of effortless creativity🚀 Free to try at: tinyurl.com/text-2-3D

@Scobleizer you have to check out what encore is doing with AI generated AR!

this feature is already live on the @ClapForEncore app! even better, you can spawn it with voice.

In honor of the big event in 🇬🇧, we made an AI visualizer of “Princess Diana” by @icespicee_ and @NICKIMINAJ using only the song lyrics. To sign up: lyrics2.ai