myhotnews

updated 0m ago

top links from the past 8h · ranked by relative velocity

  1. 1
    This is concerning. For the first time, a Chinese model Kimi K3 has taken #1 on the Frontend Code Arena and is scoring at or near the frontier on other benchmarks. Meanwhile America is tying itself in knots: politicians and bureaucrats are banning new data centers, piling on s
    AI 3h ago 253/hr +7.7σ
  2. 2
    Evidence of inconsistencies in evaluation process and selection of winners || 230 Comments
    HN 3h ago 59/hr +6.0σ
  3. 3
    US TSUNAMI WARNING SYSTEM- TSUNAMI THREAT AFTER MAGNITUDE 7.4 EARTHQUAKE NEAR THE COAST OF CHIAPAS MEXICO
    Fin 24m ago 138/hr +4.6σ
  4. 4
    IRAN WARNS U.S. FORCES Iran’s Revolutionary Guard said it is closely monitoring U.S. military movements in the region, warning that American forces are approaching the “zero hour” for potential operations against CENTCOM naval units. The statement ended with a stark message: “Wa
    Fin 41m ago 126/hr +4.2σ
  5. 5
    AWS: Inaccurate Estimated Billing Data – $1.7 billion || 227 Comments
    HN 5h ago 40/hr +3.9σ
  6. 6
    Apple targets dozens of OpenAI employees with legal letters || 133 Comments
    HN 3h ago 40/hr +3.9σ
  7. 7
    Mozilla: The state of open source AI || 29 Comments
    HN 52m ago 33/hr +3.2σ
  8. 8
    Kimi K3 may be an important inflection point for AI. Potentially negative for Anthropic and OpenAI while being net positive for essentially every other company in the world. I mean that very literally. Although the real “Sputnik moment” would be an open-source frontier model that
    Fin 1h ago 87/hr +2.7σ
  9. 9
    First atmosphere found on Earth-like planet in habitable zone of distant star || 26 Comments
    HN 1h ago 20/hr +1.8σ
  10. 10
    $SPCX - SPACEX PLUNGES FROM PEAK SpaceX shares fell 4.6% on Friday, extending a sharp slide that has erased more than $1 trillion in market value since June. The stock is now below its IPO price after a Starship launch was aborted because of an engine issue. Despite the setback,
    Fin 1h ago 61/hr +1.8σ
  11. 11
    SOCCER WORLD CUP FINAL HALFTIME TO LAST MORE THAN 15 MINUTES, BROADCASTING SOURCES SAY
    Fin 1h ago 58/hr +1.6σ
  12. 12
    Ask HN: Any AWS billing issues known? Amazon forecast of 3 billion dollars || 87 Comments
    HN 4h ago 19/hr +1.6σ
  13. 13
    CHIP STOCKS HIT HARD U.S. semiconductor stocks are heading for their worst week in over 15 months, with the Philadelphia Semiconductor Index down 12.5%. Investors are questioning the strength of the AI-driven rally after China’s Moonshot unveiled a powerful open AI model. Nvidia
    Fin 1h ago 56/hr +1.6σ
  14. 14
    U.S. MILITARY: ON JULY 16, U.S. FORCES SUCCESSFULLY DESTROYED THE CHAH BAHAR SHAHID KALANTARI PORT SURVEILLANCE TOWER
    Fin 52m ago 51/hr +1.4σ
  15. 15
    If the ingredients to strong AI models are just: Nvidia chips and good engineers, it's not really weird that Chinese open models are near parity. But why hasn't OpenAI's dominant market share translated into reams of unique data, that it can turn into compounding advantage?
    Fin 49m ago 43/hr +1.1σ
  16. 16
    Multi-Primary Color Display Emerges as Next-Gen Color Reproduction Technology || 33 Comments
    HN 2h ago 14/hr +1.1σ
  17. 17
    Newly released footage of US forces destroying Iran’s Chabahar port control tower this morning. At least 3 munitions were dropped on the base of the tower, knocking it over.
    Fin 34m ago 36/hr +0.8σ
  18. 18
    Trump Media to sell instant access to 'market-moving' social posts || 81 Comments
    HN 7h ago 11/hr +0.8σ
  19. 19
    I Owe My Life to the Commodore 64 || 54 Comments
    HN 5h ago 9.5/hr +0.6σ
  20. 20
    A lot of swift conclusions are being drawn about Kimi K3 based on fairly saturated benchmarks and ELOs, rather than actually testing it on very hard problems. The AI frontier has already moved so far that a good model that is a still months behind looks like the future to many.
    AI 37m ago 27/hr +0.6σ
  21. 21
    Minikotlin || 22 Comments
    HN 2h ago 8.6/hr +0.5σ
  22. 22
    Claude Code: Anatomy of a Misfeature || 8 Comments
    HN 56m ago 8.4/hr +0.5σ
  23. 23
    Three ways people respond to a problem (other than solving it) || 11 Comments
    HN 1h ago 7.9/hr +0.4σ
  24. 24
    Ask HN: Do you say please and thank you to your LLMs? || 46 Comments
    HN 5h ago 7.7/hr +0.4σ
  25. 25
    Kimi K3 is really, really good.
    AI 4h ago 21/hr +0.4σ
  26. 26
    "Early in the rollout of the technology, employees also faced restrictions on using Gemini to write or analyze software over concerns that proprietary code could leak into the AI model’s training data, they said." ... concerns about their OWN code being trained on?
    AI 30m ago 17/hr +0.3σ
  27. 27
    Show HN: Watch bots interact with an SSH honeypot in real time || 8 Comments
    HN 1h ago 6.1/hr +0.3σ
  28. 28
    I'm not supposed to pull back the curtain on this sort of thing, but you deserve to see what it's like to be a Big Giant FinTwit Celebrity. Sometimes you get amazing offers like the one I just got via DM.
    Fin 1h ago 20/hr +0.2σ
  29. 29
    Confirmed just press cmd+cmd
    AI 14m ago 14/hr +0.2σ
  30. 30
    VulnHunter: Capital One's agentic AI code security tool || 14 Comments
    HN 2h ago 5.2/hr +0.2σ
  31. 31
    It’ll be funny if consulting (helping companies set up AI on their computers) ends up being a higher margin business than selling the technology.
    Fin 1h ago 17/hr +0.1σ
  32. 32
    /grill-me and /grill-with-docs are changing No more "one question at a time" - now it asks questions in rounds Faster, less token spend, but still keeps dependencies between questions clear. Shipping in v1.2
    AI 4h ago 12/hr +0.1σ
  33. 33
    Spend the afternoon controlling Claude Code with an XBOX controller and a microphone Extremely fun, and pretty productive
    AI 1h ago 11/hr +0.1σ
  34. 34
    Great research paper on optimizing harnesses. (bookmark it) There is a lot of alpha in building a harness. And you don't need much to keep them optimized. This paper argues you can do this effectively using the harness own executions. The harness is the external control layer || arxiv.org
    AI 1h ago 11/hr +0.1σ
  35. 35
    K3 > GLM 5.2?
    AI 3h ago 11/hr +0.1σ
  36. 36
    me at the photoshoot: i dont know what to do with my hands... photographer: can someone give jason his laptop back? let him hold his laptop to help him relax?!
    AI 40m ago 10/hr +0.1σ
  37. 37
    we made a similar decision in the v2 plugin api had all these fancy reactive apis i tried but then traded it off for an imperative .reload() verbose apis are fine now
    AI 28m ago 10/hr +0.0σ
  38. 38
    SpaceX declared a military target by Iran || 2 Comments
    HN 24m ago 4.0/hr +0.0σ
  39. 39
    I appreciate @dhh's thoughts, but I disagree on where Europe's problems are, and from where I'm sitting it's not immigration. Thus I wrote a response to his post. dark.ronacher.eu/2026/7/17/l… || dark.ronacher.eu
    AI 3h ago 9.1/hr +0.0σ
  40. 40
    Right now it feels like the single biggest competitive advantage an AI lab could have is making it abundantly clear whether and how they will train models on your data I pay pretty close attention to this and I couldn't confidently summarize the policies for ANY of the lead labs
    AI 26m ago 8.0/hr −0.0σ
  41. 41
    guess who is advocating for acceleration? - our former "AI czar" A lot of people on twitter got one-shot by Kimi-K3 into believing the gap is now magically gone, what do you think happens to clueless politicians? the race has begun, the US will soon go into hyperdrive
    AI 2h ago 8.0/hr −0.0σ
  42. 42
    Was going to take today off and go to New York but I slept in and missed my flight cause I was making a movie in iMovie…
    AI 13m ago 6.0/hr −0.1σ
  43. 43
    Kimi K3 ranks second overall on the Debate Benchmark, trailing only Claude Fable 5! However, it is much more expensive to run than Kimi K2.6.
    AI 15m ago 6.0/hr −0.1σ
  44. 44
    Kimi is, as I have been saying, a very good model. But it is not a DeepSeek r1 moment, in that it is roughly where I would expect on the curve rather than an unexpected leap. It will be treated as a DeepSeek moment for a variety of reasons especially as more people hear about it.
    AI 25m ago 6.0/hr −0.1σ
  45. 45
    sigh
    AI 1h ago 6.0/hr −0.1σ
  46. 46
    Knowledge Atlas aka Z.AI aka Zhipu aka GLM5.2 plunged 28% last night, and is now down 63% since its June peak, when it was all anyone was talking about on here. || z.ai
    Fin 33m ago 11/hr −0.1σ
  47. 47
    A Road to Lisp: Which Lisp || 4 Comments
    HN 1h ago 2.7/hr −0.1σ
  48. 48
    Don’t sleep on open models in open harnesses! Kimi K3 in Pi means: - native multimodal support - long horizon, self-evolving workflows - 6.3x faster million-token context There’s never been a better time to be building in the open! Welcome @Kimi_Moonshot K3 to Pi
    AI 2h ago 4.8/hr −0.1σ
  49. 49
    New flow I've been enjoying 1. "/grill-with-docs <issue description>" 2. "Oh damn, this is way bigger than I expected" 3. "/wayfinder make a map of this" 4. Continue happily on
    AI 7h ago 4.8/hr −0.1σ
  50. 50
    it's worth noting that due to rising costs of pretraining, the frontier isn't *that* accessible. training a 5T-A100B model on 50T tokens on GB300s costs like $10 million for the hero run. double it for RL. there's only a few thousand companies in the world that could justify it
    AI 6h ago 4.3/hr −0.1σ