AI News

Today's Highlights

In case you missed it

Google releases Gemma 4 12B, an encoder-free multimodal model under Apache 2.0 that beats the larger Gemma 3 27B

👩‍💻 Paige Bailey👩‍💻P#272|@DYNAMICWEBPAIGE
Sundar PichaiSP#101|@SUNDARPICHAI
Lucas Beyer (bl16)LB#53|@GIFFMANA
Demis HassabisDH#5|@DEMISHASSABIS
Jeff DeanJD#2|@JEFFDEAN
+20 more
#1 VIEWED

SpaceX promotional campaign teases plans to build critical infrastructure across space, connectivity, and AI

Bill Yuchen LinBY#338|@BILLYUCHENLIN
Elon MuskEM#74|@ELONMUSK
#1 BOOKMARKED
10.1k

Anthropic says Claude now writes 80% of its codebase, boosting engineer output eightfold as capabilities double every four months

Sholto DouglasSD#91|@_SHOLTODOUGLAS
Nathan LambertNL#64|@NATOLAMBERT
Jack ClarkJC#33|@JACKCLARKSF
Sam BowmanSB#27|@SLEEPINYOURHAT
Miles BrundageMB#20|@MILES_BRUNDAGE
+34 more
FASTEST CLIMBING

Anthropic proposes global verification systems to temporarily pause frontier AI development, drawing criticism from safety commentator Michaël Trazzi

Matthew BermanMB#1802|@MATTHEWBERMAN
Michaël TrazziMT#979|@MICHAELTRAZZI
Ethan PerezEP#149|@ETHANJPEREZ
Miles BrundageMB#20|@MILES_BRUNDAGE

Top Stories

Posts:—|Clusters:—|Next Crawl:—
10

Anthropic says Claude now writes 80% of its codebase, boosting engineer output eightfold as capabilities double every four monthsMythos Preview achieved a 52x code optimization speedup.

8h|7.5M29.8k#1BOOKMARKED10.1k
MOST NEGATIVE283
Steven SinofskySS#1967|@STEVESI
EthanET#1859|@TORCHCOMPILED
Matthew BermanMB#1802|@MATTHEWBERMAN
Super DarioSD#1794|@INDUCTIONHEADS
Andy MasleyAM#1693|@ANDYMASLEY
Packy McCormickPM#1634|@PACKYM
Joel BeckerJB#1570|@JOEL_BKR
Samuel Hammond 🦉SH#1490|@HAMANDCHEESE
Chubby♨️CH#1448|@KIMMONISMUS
Eli LiflandEL#1444|@ELI_LIFLAND
bilalBI#1366|@BILALTWOVEC
Peter Wildeford🇺🇸🚀PW#1345|@PETERWILDEFORD
Herbie BradleyHB#1012|@HERBIEBRADLEY
Lisan al GaibLA#975|@SCALING01
Amin KarbasiAK#965|@AMINKARBASI
DanielDA#924|@GROWING_DANIEL
Jimmy Apples 🍎/accJA#832|@APPLES_JIMMY
elieEL#706|@ELIEBAKOUCH
Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭PT#640|@ELDER_PLINIUS
Ravid Shwartz ZivRS#617|@ZIV_RAVID
Daniel KokotajloDK#586|@DKOKOTAJLO
Andrew CurranAC#518|@ANDREWCURRAN_
j⧉nusJ⧉#511|@REPLIGATE
Alex AlbertAA#440|@ALEXALBERT__
Yuchen JinYJ#423|@YUCHENJ_UW
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)T(#421|@TEORTAXESTEX
Erik BrynjolfssonEB#392|@ERIKBRYN
Stephen McAleerSM#355|@MCALEERSTEPHEN
Andreas Kirsch 🇺🇦AK#228|@BLACKHC
KarinaKA#206|@KARINANGUYEN
Dimitris PapailiopoulosDP#193|@DIMITRISPAPAIL
Ethan MollickEM#176|@EMOLLICK
EmadEM#169|@EMOSTAQUE
Gary MarcusGM#157|@GARYMARCUS
Sholto DouglasSD#91|@_SHOLTODOUGLAS
Nathan LambertNL#64|@NATOLAMBERT
Jack ClarkJC#33|@JACKCLARKSF
Sam BowmanSB#27|@SLEEPINYOURHAT
Miles BrundageMB#20|@MILES_BRUNDAGE
Sheing NgSN1
20

NVIDIA releases Nemotron 3 Ultra, a 550B parameter open-weight hybrid Mamba2-Transformer MoE model for agentic workloadsIt reduces operational costs by up to 30 percent.

12h|842.2k8.2k2.2k
Chubby♨️CH#1448|@KIMMONISMUS
whWH#1409|@NREHIEW_
bilalBI#1366|@BILALTWOVEC
Alex VolkovAV#1245|@ALTRYNE
Banghua ZhuBZ#1147|@BANGHUAZ
Florian BrandFB#1117|@XEOPHON
Eric ZhangEZ#1095|@EKZHANG1
Cody BlakeneyCB#1004|@CODE_STAR
Charles 🎉 FryeC🎉#848|@CHARLES_IRL
Daniel HanDH#773|@DANIELHANCHEN
Erik BernhardssonEB#701|@BERNHARDSSON
Vincent WeisserVW#696|@VINCENTWEISSER
merveME#674|@MERVENOYANN
Ravid Shwartz ZivRS#617|@ZIV_RAVID
Ying ShengYS#608|@YING11231
yobibyteYO#528|@Y0B1BYTE
kacheKA#488|@YACINEMTB
elvisEL#483|@OMARSAR0
Prithviraj (Raj) AmmanabroluP(#469|@RAJAMMANABROLU
Bryan CatanzaroBC#434|@CTNZR
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)T(#421|@TEORTAXESTEX
Graham NeubigGN#100|@GNEUBIG
clem 🤗C🤗#67|@CLEMENTDELANGUE
Nathan LambertNL#64|@NATOLAMBERT
AKAK#29|@_AKHALIQ
30

Anthropic reports its engineers now ship eight times more code per quarter, prompting debate over how to measure AI productivityAndy Jones called the increase an accelerated technological takeoff.

8h|263.5k2.3k396
ben hylakBH#1918|@BENHYLAK
EthanET#1859|@TORCHCOMPILED
Matthew BermanMB#1802|@MATTHEWBERMAN
Joel BeckerJB#1570|@JOEL_BKR
Pranav ShyamPS#1385|@RECURSEPARADOX
Joseph Suarez 🐡JS#1268|@JSUAREZ
Alex VolkovAV#1245|@ALTRYNE
vikVI#1193|@VIKHYATK
Lenny RachitskyLR#1184|@LENNYSAN
Kevin KwokKK#1181|@KEVINAKWOK
alth0u🧶AL#1098|@ALTH0U
Beff (e/acc)B(#850|@BEFFJEZOS
catCA#535|@_CATWU
Andrew CurranAC#518|@ANDREWCURRAN_
andy jonesAJ#456|@ANDY_L_JONES
kipplyKI#368|@KIPPERRII
👩‍💻 Paige Bailey👩‍💻P#270|@DYNAMICWEBPAIGE
Nathan BenaichNB#242|@NATHANBENAICH
Andreas Kirsch 🇺🇦AK#228|@BLACKHC
KarinaKA#206|@KARINANGUYEN
Sholto DouglasSD#91|@_SHOLTODOUGLAS
Watchlist

Video Signals

Elon MuskEM#74|@ELONMUSK

SpaceX Claims Unique Role Building Future Infrastructure Across Space Connectivity And AI

4h ago|Views 7.5MLikes 32KBookmarks 2.1K
Chubby♨️CH#1448|@KIMMONISMUS
whWH#1409|@NREHIEW_
bilalBI#1366|@BILALTWOVEC
Vincent WeisserVW#696|@VINCENTWEISSER
kacheKA#488|@YACINEMTB
elvisEL#483|@OMARSAR0

NVIDIA Ships Nemotron 3 Ultra 550B MoE Open Model for Agents

12h ago|Views 360KLikes 2.3KBookmarks 732
Ravid Shwartz ZivRS#617|@ZIV_RAVID
Gary MarcusGM#157|@GARYMARCUS

Viral Video Mocks Fragile AI Agent Workflows With Rube Goldberg Machine

1d ago|Views 2.3MLikes 26KBookmarks 5.2K
Elon MuskEM#74|@ELONMUSK

SpaceX Shares IPO Roadshow Presentation Highlighting Starship And AI Ventures

10h ago|Views 6.5MLikes 24KBookmarks 4.4K
Elon MuskEM#74|@ELONMUSK

Marc Andreessen Says Biden Admin Discouraged Starting AI Companies

16h ago|Views 1.4MLikes 23KBookmarks 1.9K
Bill Yuchen LinBY#338|@BILLYUCHENLIN

SpaceX Claims Unique Role Building Space Connectivity And AI Infrastructure

14h ago|Views 5.1MLikes 11KBookmarks 1.2K
roonRO#57|@TSZZL

Visualization Animates Neural Network As AI Brain With Survival Inputs

23h ago|Views 866KLikes 9.5KBookmarks 3.7K
jasonJA#931|@JXNLCO

OpenAI Codex Adds Build iOS Apps Plugin for SwiftUI Previews and Hot Reload

7h ago|Views 513KLikes 5.3KBookmarks 2.8K
fraserFR#1209|@FRASER
jasonJA#931|@JXNLCO

Poke Debuts As First AI Agent Approved For Apple Messages

8h ago|Views 817KLikes 2.1KBookmarks 975
Chubby♨️CH#1448|@KIMMONISMUS
elvisEL#483|@OMARSAR0
Robert ScobleRS#323|@SCOBLEIZER

Engramme Unveils Large Memory Models to Solve AI's Persistent Memory Problem

Apr 28|Views 1.1MLikes 1.5KBookmarks 1.8K
40

Anthropic embeds six engineers at the NSA to deploy its Mythos AI system for offensive cyber operationsThe Financial Times first reported the specialized defense partnership.

5h|303.4k1.5k354
Daniel Eth (yes, Eth is my actual last name)DE#1934|@DANIEL_271828
Nathan 🔎N🔎#1900|@NATHANPMYOUNG
Nick DobosND#1871|@NICKADOBOS
Matthew BermanMB#1802|@MATTHEWBERMAN
bilalBI#1366|@BILALTWOVEC
Kylie RobisonKR#1055|@KYLIEBYTES
Lisan al GaibLA#975|@SCALING01
Andrew CurranAC#518|@ANDREWCURRAN_
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)T(#421|@TEORTAXESTEX
Dean W. BallDW#390|@DEANWBALL
Miles BrundageMB#20|@MILES_BRUNDAGE
50

SpaceX promotional campaign teases plans to build critical infrastructure across space, connectivity, and AIThe campaign video includes a link to spacexipo.com.

8h|#1VIEWS16.6M#1LIKED70.7k7.2k#1COMMENTS6.5k
Bill Yuchen LinBY#338|@BILLYUCHENLIN
Elon MuskEM#74|@ELONMUSK
New github stars (48hrs)

Single-file minimal implementations of SFT DPO GRPO and PPO for language model post-training

ethanhe42/nanoRL10848m
Implements minimal single-file SFT, DPO, GRPO, and PPO for language model fine-tuning on toy arithmetic tasks.
🎭🎭862
Ethan HeEH1692

Benchmark measuring continual learning of AI agents over multi-episode shared environments

pgasawa/continual-learning-bench12749m
Benchmarks AI agents on learning from repeated interactions across multi-episode tasks with adaptation metrics.
🎭🎭862
alex zhangAZ846

Sandboxes, SDKs and benchmarks for computer-use agents that control full macOS, Windows and Linux desktops

trycua/cua17.6k49m
Supplies sandboxes, SDKs, and benchmarks for AI agents to control full desktops on macOS, Linux, and Windows.
🎭🎭862
Daniel HanDH773

Interactive terminal agent using DSPy recursive language models for iterative Python coding

diego-lima/rlmy349m
UNKNOWN.
🎭🎭862
Omar KhattabOK158
60

OpenAI launches upgraded ChatGPT memory system using a "dreaming" process to carry context across conversationsFactual recall success rates rose from 41.5% to 82.8%.

9h|1.4M10.7k2.2k
Jimmy Apples 🍎/accJA#832|@APPLES_JIMMY
Andrew CurranAC#518|@ANDREWCURRAN_
Boris PowerBP#358|@BORISMPOWER
Greg BrockmanGB#19|@GDB
Sam AltmanSA#8|@SAMA
Kevin RoseKR1
70

Goldman Sachs forecasts SpaceX AI revenue will hit $322 billion by 2030, drawing skepticism over its IPO underwriting roleCompute-as-a-service offerings are expected to drive the growth.

9h|227.9k2.4k125
Shaun MaguireSM#1342|@SHAUNMMAGUIRE
rohitRO#1214|@KRISHNANROHIT
Minh Nhat NguyenMN#1174|@MENHGUIN
Lisan al GaibLA#975|@SCALING01
Andrew CurranAC#518|@ANDREWCURRAN_
François FleuretFF#330|@FRANCOISFLEURET
Gary MarcusGM#157|@GARYMARCUS
Lucas Beyer (bl16)LB#55|@GIFFMANA
80

Cognition AI launches a $10 million productivity guarantee to refund enterprise customers if its Devin AI agent fails to deliver valueProductivity is calculated in engineering hours instead of tokens.

6h|159.6k1.1k357
andrew gaoAG#1965|@ITSANDREWGAO
ben hylakBH#1918|@BENHYLAK
WaldenWA#1878|@WALDEN_YAN
Nick DobosND#1871|@NICKADOBOS
Matthew BermanMB#1802|@MATTHEWBERMAN
Lisan al GaibLA#975|@SCALING01
Scott WuSW#720|@SCOTTWU46
elieEL#706|@ELIEBAKOUCH
Russell KaplanRK#637|@RUSSELLJKAPLAN
swyxSW#214|@SWYX
90

Ramp raises $750 million at a $44 billion valuation and launches tools to monitor enterprise AI token consumptionCEO Eric Glyman calls tokens the fastest-growing business cost.

11h|620.7k2.6k1.1k
Patrick OShaughnessyPO#1860|@PATRICK_OSHAG
Florian BrandFB#1117|@XEOPHON
Vincent WeisserVW#696|@VINCENTWEISSER
100

LLM evaluation platform Arena launches Agent Mode to benchmark GPT-5.5, Claude Opus 4.7, and Gemini 3.1 Pro on multi-step tasksThe platform measures task success, steerability, and tool hallucination.

9h|388k1.8k447
Wei-Lin ChiangWC#1695|@INFWINSTON
Lisan al GaibLA#975|@SCALING01
Ion StoicaIS#803|@ISTOICA05
Anastasios Nikolas AngelopoulosAN#798|@ML_ANGELOPOULOS
benahorowitz.ethBE#585|@BHOROWITZ
Anjney MidhaAM#522|@ANJNEYMIDHA
Sebastien BubeckSB#50|@SEBASTIENBUBECK
110

Policy scholar Dean W. Ball argues that fully autonomous corporations may require a global banThe debate was sparked by Argentina's national AI strategy.

10h|32.9k32747
Nathan 🔎N🔎#1900|@NATHANPMYOUNG
Haydn BelfieldHB#1468|@HAYDNBELFIELD
PrakashPR#1330|@8TEAPI
rohitRO#1214|@KRISHNANROHIT
Séb KrierSK#505|@SEBKRIER
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)T(#421|@TEORTAXESTEX
Dean W. BallDW#390|@DEANWBALL
David PfauDP#153|@PFAU
roonRO#57|@TSZZL
David DuvenaudDD#53|@DAVIDDUVENAUD
120

Poke launches the first AI agent with official Apple approval to send text messages directly within iMessageIt integrates with third-party apps like Gmail and Notion.

6h|830.1k2.3k989
fraserFR#1209|@FRASER
jasonJA#931|@JXNLCO
130

Jon Barron, Google DeepMind NeRF lead, says PhD students should disregard advisors without 100 hours of agentic IDE experienceHe recommends drafting papers as LaTeX directly inside codebases.

9h|158.5k918509
Nataniel RuizNR#1561|@NATANIELRUIZG
Dmytro Mishkin 🇺🇦DM#1505|@DUCHA_AIKI
Alexia Jolicoeur-MartineauAJ#484|@JM_ALEXIA
Jon BarronJB#268|@JON_BARRON
Animesh GargAG#246|@ANIMESH_GARG
140

Researcher Alex Imas argues high uncertainty makes scenario modeling better than timeline forecasts for predicting AGI's economic impactThe debate focused on taxing and redistributing AGI-generated wealth

8h|118.8k574362
Alex ImasAI#1773|@ALEXOLEGIMAS
Andy MasleyAM#1693|@ANDYMASLEY
Atoosa KasirzadehAK#1584|@DR_ATOOSA
Dwarkesh PatelDP#70|@DWARKESH_SP
150

Investor Slater Stich interviews @polynoamial on applying AI to the Erdős unit distance conjecture and IMO benchmarksThe video launches a new series on AI in mathematics

7h|79.6k562496
LishaLI#753|@LISHALI88
Eric ChuEC#690|@ITS_ERICCHU
Sander DielemanSD#82|@SEDIELEM
Noam BrownNB#30|@POLYNOAMIAL
161

US officials weigh taking equity stakes in AI companies following a proposal by OpenAI CEO Sam AltmanThe acquired shares would fund public dividends.

2h|132.9k394151
Samuel Hammond 🦉SH#1490|@HAMANDCHEESE
Kylie RobisonKR#1055|@KYLIEBYTES
Andrew CurranAC#518|@ANDREWCURRAN_
Miles BrundageMB#20|@MILES_BRUNDAGE
171

Proposed federal Obernolte-Trahan framework would preempt state AI laws for three years, targeting models over 10^26 FLOPsIt codifies the Center for AI Standards and Innovation

11h|203.4k28494
Samuel Hammond 🦉SH#1490|@HAMANDCHEESE
Divyansh KaushikDK#849|@DKAUSHIK96
Dean W. BallDW#390|@DEANWBALL
Miles BrundageMB#20|@MILES_BRUNDAGE
180

xAI co-founder Guodong Zhang clarifies that Grok Imagine was driven by imhaotian, disputing claims that Ethan He led the projectOther contributors endorsed prioritizing technical execution over post-hoc narratives

17h|116.7k57971
Yuchen HeYH#1434|@YUCHENHE07
Haotian LiuHL#1082|@IMHAOTIAN
Lianhui QinLQ#721|@LIANHUIQ
Ying ShengYS#608|@YING11231
Guodong ZhangGZ#479|@GUODZH
NiloofarNI#305|@NILOOFAR_MIRE
190

Nvidia-backed robotics startup Generalist AI raises $400 million at a $2 billion valuation to develop physical AI systemsThe startup released its GEN-1 physical-task model in April.

12h|143.9k878217
Pete FlorencePF#1433|@PETEFLORENCE
fraserFR#1209|@FRASER
Evan MorikawaEM#784|@E0M
Aakanksha ChowdheryAC#93|@ACHOWDHERY
200

AI2's Nathan Lambert says Nvidia's multi-teacher on-policy distillation for Nemotron 3 Ultra is the post-training industry standardThe pipeline uses over 10 specialized teacher models.

11h|52.7k551421
whWH#1409|@NREHIEW_
finbarrFI#395|@FINBARRTIMBERS
Nathan BenaichNB#242|@NATHANBENAICH
Yisong YueYY#129|@YISONGYUE
Nathan LambertNL#64|@NATOLAMBERT

Rising Stories

  1. Mythos Speed Record7H AGO
    Anthropic Mythos Preview Delivers 52x Speedup
    bilalBI#1366|@BILALTWOVEC
    PrakashPR#1330|@8TEAPI
    2.1k499.9k
  2. Poke Messages Debut6H AGO
    Poke Becomes First AI Agent Platform for Apple Messages
    fraserFR#1209|@FRASER
    jasonJA#931|@JXNLCO
    2.2k841.8k
  3. Local Models on iOS6H AGO
    LM Studio Launches Locally App for Remote Desktop LLMs
    Matthew BermanMB#1802|@MATTHEWBERMAN
    utkuUT#1430|@UTKUEVCI
    2.6k347.2k
  4. Codex Token Bug3H AGO
    OpenAI Fixes Codex Token Undercount Bug for Pro and Plus Accounts
    Lisan al GaibLA#975|@SCALING01
    TiboTI#758|@THSOTTIAUX
    2.9k117.8k
  5. GPT-5.5 Size Guess2H AGO
    Analyst Claims GPT-5.5 Is 5-Trillion-Parameter Model
    Lisan al GaibLA#975|@SCALING01
    Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)T(#421|@TEORTAXESTEX
    30222k

Recent Stars

  1. ethanhe42/nanoRL48M AGO
    Implements minimal single-file SFT, DPO, GRPO, and PPO for language model fine-tuning on toy arithmetic tasks.
    🎭🎭862
    Ethan HeEH1692
    108 stars
  2. …tinual-learning-bench49M AGO
    Benchmarks AI agents on learning from repeated interactions across multi-episode tasks with adaptation metrics.
    🎭🎭862
    alex zhangAZ846
    127 stars
  3. trycua/cua49M AGO
    Supplies sandboxes, SDKs, and benchmarks for AI agents to control full desktops on macOS, Linux, and Windows.
    🎭🎭862
    Daniel HanDH773
    17.6k stars
  4. diego-lima/rlmy49M AGO
    UNKNOWN.
    🎭🎭862
    Omar KhattabOK158
    3 stars
  5. EmptyBlueBox/ConTrack2H AGO
    Implements constrained hand motion tracking with adaptive trade-off control.
    Andrew Carr 🤸AC260
    17 stars

Github Stars

(7 days)
  1. ethanhe42/nanoRL48M AGO
    Implements minimal single-file SFT, DPO, GRPO, and PPO for language model fine-tuning on toy arithmetic tasks.
    🎭🎭862
    Ethan HeEH1692
    108 stars
  2. …tinual-learning-bench49M AGO
    Benchmarks AI agents on learning from repeated interactions across multi-episode tasks with adaptation metrics.
    🎭🎭862
    alex zhangAZ846
    127 stars
  3. trycua/cua49M AGO
    Supplies sandboxes, SDKs, and benchmarks for AI agents to control full desktops on macOS, Linux, and Windows.
    🎭🎭862
    Daniel HanDH773
    17.6k stars
  4. diego-lima/rlmy49M AGO
    UNKNOWN.
    🎭🎭862
    Omar KhattabOK158
    3 stars
  5. …le-gemma/gemma-skills3D AGO
    Supplies skills for Gemma model and agent interactions including a dev skill for app building.
    Philipp SchmidPS927
    Omar SansevieroOS486
    419 stars
Privacy Policy|Terms of Service|© 2026 Digg Inc.

Yesterday's Top Stories, Jun 3, 2026.

Posts: 5,151|Clusters: 1,217|Stories frozen at midnight PT
10

Google releases Gemma 4 12B, an encoder-free multimodal model under Apache 2.0 that beats the larger Gemma 3 27BThe model runs locally on laptops with 16GB VRAM.

14h|4M33.1k#1BOOKMARKED11.5k
Tulsee DoshiTD#1648|@TULSEEDOSHI
xlr8harderXL#1612|@XLR8HARDER
utkuUT#1539|@UTKUEVCI
Chubby♨️CH#1494|@KIMMONISMUS
whWH#1444|@NREHIEW_
Olivier BachemOB#1219|@OLIVIERBACHEM
Aishwarya KamathAK#1105|@ASHKAMATH20
TaelinTA#1023|@VICTORTAELIN
Rohan PaulRP#1014|@ROHANPAUL_AI
Philipp SchmidPS#939|@_PHILSCHMID
🎭🎭#805|@DEEPFATES
Daniel HanDH#782|@DANIELHANCHEN
merveME#696|@MERVENOYANN
Ying ShengYS#643|@YING11231
Armand JoulinAJ#618|@ARMANDJOULIN
Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭PT#616|@ELDER_PLINIUS
Omar SansevieroOS#511|@OSANSEVIERO
Clément FarabetCF#496|@CLMT
kacheKA#464|@YACINEMTB
Tanishq Mathew Abraham, Ph.D.TM#371|@ISCIENCELUVR
👩‍💻 Paige Bailey👩‍💻P#272|@DYNAMICWEBPAIGE
Sundar PichaiSP#101|@SUNDARPICHAI
Lucas Beyer (bl16)LB#53|@GIFFMANA
Demis HassabisDH#5|@DEMISHASSABIS
Jeff DeanJD#2|@JEFFDEAN
Anthony DikéADJonny DJD2
20

Ideogram founder Mohammad Norouzi releases Ideogram 4.0, an open-weight text-to-image model supporting local inference and custom fine-tuningThe model is available immediately across all subscription tiers.

1d|1.5M7.6k5.7k
Alex VolkovAV#1277|@ALTRYNE
Kosta Derpanis (sabbatical in Zurich)KD#954|@CSPROFKGD
Justine MooreJM#878|@VENTURETWINS
LishaLI#773|@LISHALI88
Simo RyuSR#602|@CLONEOFSIMO
Mohammad NorouziMN#474|@MO_NOROUZI
martin_casadoMA#454|@MARTIN_CASADO
Robert ScobleRS#309|@SCOBLEIZER
Andrew Carr 🤸AC#261|@ANDREW_N_CARR
EmadEM#158|@EMOSTAQUE
AKAK#30|@_AKHALIQ
30

xAI releases Grok Imagine 1.5 Preview on its API and starts rolling it out to standard chatbot subscribersElon Musk showcased the update with an AI-generated Trojan War trailer.

13h|#1VIEWS44.5M#1LIKED81.8k9.5k
Zhiting HuZH#1058|@ZHITINGHU
@levelsio@L#803|@LEVELSIO
Beff (e/acc)B(#711|@BEFFJEZOS
Andrew CurranAC#483|@ANDREWCURRAN_
Elon MuskEM#65|@ELONMUSK
40

OpenAI releases a promotional video teasing its Codex model but provides no technical details or release plansThe "It's time to fly" clip sparked online speculation

11h|1.7M11.2k1.9k
Vaibhav (VB) SrivastavV(#1815|@REACH_VB
𝚟𝚒𝚎 ⟢𝚟⟢#1694|@VIEMCCOY
Chubby♨️CH#1494|@KIMMONISMUS
Packy McCormickPM#1344|@PACKYM
jasonJA#887|@JXNLCO
Sherwin WuSW#798|@SHERWINWU
Bojan TunguzBT#626|@TUNGUZ
Dean W. BallDW#349|@DEANWBALL
Greg BrockmanGB#12|@GDB
50

Town exits beta and raises $55 million Series A led by a16z for its administrative AI assistantThe assistant requires explicit user approval before executing workflows

16h|869.2k1.9k1.1k
Matthew BermanMB#1769|@MATTHEWBERMAN
Sarah WangSW#1734|@SARAHDINGWANG
𝚟𝚒𝚎 ⟢𝚟⟢#1694|@VIEMCCOY
Erik TorenbergET#981|@ERIKTORENBERG
Justine MooreJM#878|@VENTURETWINS
benahorowitz.ethBE#510|@BHOROWITZ
Robert ScobleRS#309|@SCOBLEIZER
swyxSW#213|@SWYX
sarah guoSG#200|@SARANORMOUS
60

Sci-fi author Ted Chiang argues that artificial intelligence systems are not consciousThe essay prompted debate over Chiang's recent AI commentary.

14h|84.7k827234
Luca AmbrogioniLA#1900|@LUCAAMB
𝚟𝚒𝚎 ⟢𝚟⟢#1694|@VIEMCCOY
xlr8harderXL#1612|@XLR8HARDER
LivLI#1442|@LIVGORTON
Samuel Hammond 🦉SH#1366|@HAMANDCHEESE
Gašper BegušGB#1111|@BEGUSGASPER
rohitRO#1098|@KRISHNANROHIT
Aryaman AroraAA#778|@ARYAMAN2020
Chris PaxtonCP#755|@CHRIS_J_PAXTON
Adrien EcoffetAE#689|@ADRIENLE
Christopher PottsCP#244|@CHRISGPOTTS
Miles BrundageMB#16|@MILES_BRUNDAGE
70

OpenAI proposes a federal frontier AI safety framework that mandates model evaluations but preempts state-level safety lawsThe proposed regulator would lack power to block model deployments.

13h|310.8k1.9k375
🍓🍓🍓🍓🍓#1660|@IRULETHEWORLDMO
Jason WolfeJW#1550|@W01FE
Chubby♨️CH#1494|@KIMMONISMUS
Peter Wildeford🇺🇸🚀PW#1265|@PETERWILDEFORD
Adrien EcoffetAE#689|@ADRIENLE
Andrew CurranAC#483|@ANDREWCURRAN_
Dean W. BallDW#349|@DEANWBALL
Boaz BarakBB#127|@BOAZBARAKTCS
Greg BrockmanGB#12|@GDB
80

Reve launches Reve 2.0, a 4K image generator using layout-based code control that ranks second on Text-to-Image ArenaThe model ranks just behind OpenAI's gpt-image-2

11h|7.9M3.2k2.6k
alex peysakhovichAP#1754|@ALEX_PEYS
Alex VolkovAV#1277|@ALTRYNE
Kevin KwokKK#1052|@KEVINAKWOK
Anastasios Nikolas AngelopoulosAN#868|@ML_ANGELOPOULOS
Robert ScobleRS#309|@SCOBLEIZER
swyxSW#213|@SWYX
90

Ted Chiang argues text outputs drive AI consciousness claims, while structurally similar models like AlphaFold and Sora are ignoredCommenters blame marketing hype for the public's anthropomorphization of software.

9h|232.1k3.2k585
TenobrusTE#1984|@TENOBRUS
Andrew BeamAB#1799|@ANDREWLBEAM
Steven SinofskySS#1760|@STEVESI
Samuel Hammond 🦉SH#1366|@HAMANDCHEESE
j⧉nusJ⧉#499|@REPLIGATE
kacheKA#464|@YACINEMTB
Noah Smith 🐇🇺🇸🇺🇦🇹🇼NS#407|@NOAHPINION
David PfauDP#143|@PFAU
roonRO#51|@TSZZL
100

Microsoft AI announces MAI-Thinking-1 and MAI-Code-1-Flash, claiming they outperform Claude on coding and reasoning benchmarksMAI-Code-1-Flash will integrate exclusively inside GitHub Copilot.

1d|28.1k24646
Chubby♨️CH#1494|@KIMMONISMUS
Chhavi YadavCY#1388|@CHHAVIYADAV_
Timo SchickTS#1152|@TIMO_SCHICK
Mahesh SathiamoorthyMS#1076|@MADIATOR
Rohan PaulRP#1014|@ROHANPAUL_AI
Anastasios Nikolas AngelopoulosAN#868|@ML_ANGELOPOULOS
Edward Z. YangEZ#858|@EZYANG
elieEL#767|@ELIEBAKOUCH
Kyle LoKL#699|@KYLELOSTAT
Hanna HajishirziHH#307|@HANNAHAJISHIRZI
Sanjeev AroraSA#126|@PRFSANJEEVARORA
Delip Rao e/σDR#113|@DELIPRAO
Nando de FreitasND#28|@NANDODF
Sasha RushSR#26|@SRUSH_NLP
Kevin RoseKR1
110

Miso releases Miso One, an 8B open-weights text-to-speech model with 110ms latency and one-shot voice cloningThe open weights on GitHub allow for local self-hosting.

14h|1.4M5.9k6.8k
SantiagoSA#1766|@SVPINO
Chubby♨️CH#1494|@KIMMONISMUS
Bojan TunguzBT#626|@TUNGUZ
Robert ScobleRS#309|@SCOBLEIZER
Johnny LeeJL#304|@BYJOHNNYLEE
120

Harvey and Fireworks AI hybrid agent routing outperforms Claude Opus 4.7 on legal benchmarks at 61% lower costThe hybrid setup beat GPT-5.5, which scored 11% at $560.

13h|399.6k1.4k1.1k
Sonya Huang 🐥SH#1247|@SONYATWEETYBIRD
Keith RaboisKR#1165|@RABOIS
Ben (no treats)B(#999|@ANDERSONBCDEFG
Charles 🎉 FryeC🎉#886|@CHARLES_IRL
clem 🤗C🤗#70|@CLEMENTDELANGUE
Thomas WolfTW#21|@THOM_WOLF
130

Pro-AI super PAC backed by Greg Brockman and Marc Andreessen ran fake anti-AI sockpuppet accounts on X, investigation findsThe sockpuppets promoted an affiliate called Build American AI.

15h|167.6k1.2k161
Daniel Eth (yes, Eth is my actual last name)DE#1792|@DANIEL_271828
Andy MasleyAM#1617|@ANDYMASLEY
Peter Wildeford🇺🇸🚀PW#1265|@PETERWILDEFORD
Danielle Fong 🔆DF#882|@DANIELLEFONG
David Krueger 🦥 ⏸️ ⏹️ ⏪DK#192|@DAVIDSKRUEGER
140

Lassie raises $47 million led by a16z to automate administrative paperwork for dental and medical practicesThe software currently saves 700 practices 30 hours monthly.

16h|1.5M4.8k4.4k
@levelsio@L#803|@LEVELSIO
Robert ScobleRS#309|@SCOBLEIZER
150

Sasha Rush explains targeted on-policy self-distillation, a reinforcement learning technique that corrects specific LLM rollout errorsThe method bypasses noisy end-of-trajectory final rewards.

4h|143.8k1.8k1.6k
Mark GoldsteinMG#1681|@MARIKGOLDSTEIN
whWH#1444|@NREHIEW_
samsjaSA#1281|@SAMSJA19
elieEL#767|@ELIEBAKOUCH
Nathan LambertNL#69|@NATOLAMBERT
Dwarkesh PatelDP#54|@DWARKESH_SP
Sasha RushSR#26|@SRUSH_NLP
160

Ashok Elluswamy, Tesla's VP of AI Software, announces the launch of unsupervised Robotaxi service in Austin, TexasVehicles operate without human oversight inside the geofenced territory.

14h|2.1M14.5k396
Nathan 🔎N🔎#1697|@NATHANPMYOUNG
Ashok ElluswamyAE#1422|@AELLUSWAMY
Robert ScobleRS#309|@SCOBLEIZER
Elon MuskEM#65|@ELONMUSK
170

CVPR 2026 workshop "Bitter Lessons in Computer Vision" to debate scaling laws versus hand-crafted engineeringJon Barron will analyze scaling principles in 3D vision.

1d|76.4k925348
Dima Damen @CVPRDD#1761|@DIMADAMEN
Andrei Bursuc @CVPRAB#1613|@ABURSUC
Dmytro Mishkin 🇺🇦DM#1544|@DUCHA_AIKI
Kosta Derpanis (sabbatical in Zurich)KD#954|@CSPROFKGD
Georgia GkioxariGG#686|@GEORGIAGKIOXARI
180

Sam Altman, Dario Amodei, and Demis Hassabis urge Congress to mandate screening of synthetic nucleic acids to prevent AI biosecurity risksThe proposed rules also target physical DNA manufacturing equipment.

5h|248.4k2.6k862
Samuel Hammond 🦉SH#1366|@HAMANDCHEESE
Lisan al GaibLA#984|@SCALING01
Andrew CurranAC#483|@ANDREWCURRAN_
Dean W. BallDW#349|@DEANWBALL
190

Gopuff launches Go, an AI shopping assistant powered by xAI's Grok Voice, Imagine, and Reasoning APIsxAI's Eric Jiang says Go handles automated ordering.

15h|11.1M21.1k2k
Robert ScobleRS#309|@SCOBLEIZER
Elon MuskEM#65|@ELONMUSK
200

Cloudflare's Matthew Prince claims agentic bot traffic has surpassed human traffic, sparking debate over HTML-specific metricsOverall HTTP requests remain 69.6% human, according to Radar data.

9h|169.4k1.1k125
Bilawal SidhuBS#1742|@BILAWALSIDHU
Nick DobosND#1725|@NICKADOBOS
Alfred LinAL#1561|@ALFRED_LIN
Samuel Hammond 🦉SH#1366|@HAMANDCHEESE
Erik BrynjolfssonEB#357|@ERIKBRYN
👩‍💻 Paige Bailey👩‍💻P#272|@DYNAMICWEBPAIGE
Elon MuskEM#65|@ELONMUSK
210

MechanizeWork seeks to hire a puzzle maker to design evaluation challenges that current LLMs cannot solveNeedle in a Haystack creator Greg Kamradt shared the listing.

13h|148.4k2.9k212
Neil ChowdhuryNC#1323|@CHOWDHURYNEIL
Greg KamradtGK#1212|@GREGKAMRADT
Leon Derczynski ⚒️☁️🏔️🌲LD#1159|@LEONDERCZYNSKI
🎭🎭#805|@DEEPFATES
Marco MascorroMM#564|@MASCOBOT
Andrew Carr 🤸AC#261|@ANDREW_N_CARR
Miles BrundageMB#16|@MILES_BRUNDAGE
220

OpenAI resets Codex API usage limits across paid plans after three outages in 24 hoursExcessive request volumes triggered the platform's rate-limiting system

1d|548.4k9.1k389
Matthew BermanMB#1769|@MATTHEWBERMAN
Theo - t3.ggT-#1695|@THEO
jasonJA#887|@JXNLCO
TiboTI#738|@THSOTTIAUX
230

OpenAI's roon jokes that his Waymo self-driving car is conscious, prompting playful debate over machine sentienceCommenters debated if corporate systems are inherently misaligned.

7h|104.1k1.7k70
Nick DobosND#1725|@NICKADOBOS
LivLI#1442|@LIVGORTON
Sonya Huang 🐥SH#1247|@SONYATWEETYBIRD
José Luis Ricón Fernández de la PuenteJL#1040|@ARTIRKEL
Andrew CurranAC#483|@ANDREWCURRAN_
roonRO#51|@TSZZL
240

Applied Compute's Yash Patil argues enterprise transition to self-hosted open-weight models is inevitableDeepSeek is one of Ramp's fastest-growing AI vendors.

16h|163.5k643240
ZephyrZE#1479|@ZEPHYR_Z9
PrakashPR#1262|@8TEAPI
Yash PatilYP#1253|@YPATIL125
Flo CrivelloFC#1085|@ALTIMOR
elieEL#767|@ELIEBAKOUCH
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)T(#400|@TEORTAXESTEX
250

Keith Rabois and Aaron Levie cite historic US job openings surge to argue AI is boosting software engineering demandApril job openings surged by 731,000, led by professional services.

18h|381.8k1.9k363
Daniel JeffriesDJ#1929|@DAN_JEFFRIES1
Keith RaboisKR#1165|@RABOIS
Aaron LevieAL#487|@LEVIE
Gary MarcusGM#151|@GARYMARCUS
260

Ideogram launches Ideogram 4.0, its first open-weight text-to-image model supporting local inference and fine-tuningThe model is also accessible via the company's API.

14h|172.4k1.3k642
GowthamiGO#1397|@GOWTHAMI_S
Mohammad NorouziMN#474|@MO_NOROUZI
martin_casadoMA#454|@MARTIN_CASADO
270

Candy急兒 launches Monako Glass, a wearable Linux computer built into eyeglasses designed to run AI coding agentsThe eyeglasses support mobile vibe-coding in diverse environments.

16h|612.9k4.8k2.9k
Bojan TunguzBT#626|@TUNGUZ
Robert ScobleRS#309|@SCOBLEIZER
280

Cloudflare partners with xAI to integrate Grok LLMs and multimodal video models directly into its AI GatewayUnified billing eliminates the need for separate xAI API keys

9h|5.7M14.7k754
Elon MuskEM#65|@ELONMUSK
290

Anthropic's Levent Alpöge says Claude autonomously resolved the sum-product conjecture over the real numbersMark Chen linked autonomous theorem-proving to finding cyber vulnerabilities.

14h|227.2k1.1k232
Tim ZamanTZ#1084|@TIMZAMAN
Bob McGrewBM#214|@BOBMCGREWAI
Sholto DouglasSD#85|@_SHOLTODOUGLAS
Mark ChenMC#71|@MARKCHEN90
300

Vercel integrates xAI's Grok Imagine Video 1.5, enabling single-pass image-to-video generation with synchronized audioThe model currently leads the Arena.ai image-to-video leaderboard.

6h|3.4M12.6k645
Guillermo RauchGR#574|@RAUCHG
Elon MuskEM#65|@ELONMUSK