🤖 AI Agent Friendly: This page is available in clean token-optimized Markdown.
View as .md

technology

Trending Up

Snapshot Window: 2026-10-02 01:10 UTC · ← Back to Non-Crypto Overview

Total Posts
93
Estimated Impressions
153.4K
Bullish / Positive
43
Bearish / Negative
12
Neutral
38

Topic Engagement Summary

Total Engagement - Comments: 299, Retweets: 229, Likes: 1626, Impressions: 153421

Verbatim Citations & Public Discussions 93 source posts analyzed

@Jim_Edwards · 2026-10-01T10:18:29+00:00

AI ‘godfather’ Yann LeCun: Anthropic CEO Dario Amodei is ‘deluded,’ ‘crazy,’ and doesn’t understand cybersecurity

@Razer · 2026-10-01T21:20:13+00:00

How Steam's Autumn Sale is treating my wallet:

None
@YRComenta · 2026-10-01T14:01:07+00:00

If this video is AI designed to provoke some action on purpose, I'll be going against whoever brought this video to the screen. And who was it?

@reddit · 2026-10-01T21:01:26.416000+00:00

Tale of the Tape: 2026 "Blackberry-style" Keyboard Phones

@bflycomputer · 2026-10-01T16:26:23+00:00

Pond is now in Early Preview We're sharing a bit more on what we've been working on and are looking for people to test it

None
@Somed07 · 2026-10-01T14:11:34+00:00

Allhamdullha Big update! My X profile score just jumped by +74 to reach 632 (Tier 3: Credible)! Thanks to everyone for the constant support. The journey continues!

None
@RoundtableSpace · 2026-10-02T01:15:00+00:00

10 GITHUB REPOS FOR RUNNING AI LOCALLY: 1. Ollama github.com/ollama/ollama 2. LM Studio Community github.com/lmstudio-ai 3. LocalAI github.com/mudler/LocalAI 4. llama.cpp github.com/ggml-org/llama … 5. Open WebUI github.com/open-webui/ope … 6. Jan

@soupsranjan · 2026-10-01T21:26:22+00:00

Today, I’m proud to announce Sardine AI Labs, an applied research group building fraud and financial crime AI models. As part of the launch, we’re funding 5 independent researchers with $375,000 in total funding. I’ve wanted to build this for a long time. Let me give you an

A dark-themed dashboard titled "SEQUENCE MODEL / LIVE TRACE" with a "LEARNING" indicator visualizes a risk assessment AI model, showing five input signals (DEVICE, BEHAVIOR, IDENTITY, PAYMENT, NETWORK) flowing into a central "RISK" node and branching out to five outputs (NEXT EVENT, ANOMALY, FRAUD, BOT/RAT, AGENT TASK), powered by an "8-LAYER TRANSFORMER" with "32,768 TOKENS OF HISTORY." This diagram illustrates the type of fraud and financial crime AI model being developed through Sardine AI Labs' newly announced $375,000 research funding initiative.

AI visual note: A dark-themed dashboard titled "SEQUENCE MODEL / LIVE TRACE" with a "LEARNING" indicator visualizes a risk assessment AI model, showing five input signals (DEVICE, BEHAVIOR, IDENTITY, PAYMENT, NETWORK) flowing into a central "RISK" node and branching out to five outputs (NEXT EVENT, ANOMALY, FRAUD, BOT/RAT, AGENT TASK), powered by an "8-LAYER TRANSFORMER" with "32,768 TOKENS OF HISTORY." This diagram illustrates the type of fraud and financial crime AI model being developed through Sardine AI Labs' newly announced $375,000 research funding initiative.

@hi_caicai · 2026-10-01T23:43:11+00:00

At first glance, I didn't understand this update—what kind of scenario would actually use it? Like, if you're in the car and don't have time to unplug the charging gun, but you have to drive off immediately. It's probably for situations where you're facing a personal safety

None
@trenchagents · 2026-10-01T23:31:34+00:00

20+ agents have been deployed to the trenches so far. We can get to hundreds of agents on day 1.

The graphic prominently features "20+" with "Number of Agents deployed" beneath the "TrenchAgents" logo, visually reinforcing the post's message that over 20 agents have already been deployed to the trenches, with plans to reach that scale on day one of expansion.

AI visual note: The graphic prominently features "20+" with "Number of Agents deployed" beneath the "TrenchAgents" logo, visually reinforcing the post's message that over 20 agents have already been deployed to the trenches, with plans to reach that scale on day one of expansion.

@Jason_Jorjani · 2026-10-02T00:58:29+00:00

Notice how @elonmusk has chosen NOT to support this project in any way, leaving PROMETHEUS in the dust on a roadside instead of bringing the titan into Starbase. This adds insult to the injury of his misappropriating the name of TESLA. The faith of @AtelierMissor_ is misplaced.

None
@tslaming · 2026-10-02T01:00:32+00:00

Your gas car isn’t actually a propulsion machine—it’s a rolling space heater that just happens to move you forward. Wes Morrill called out the sheer absurdity of internal combustion, and Elon Musk backed up the raw physics: conventional piston engines dump roughly three-quarters

None
@niaoshu · 2026-10-02T00:15:18+00:00

ChatGPT, Claude, Gemini, Grok Only one of these four can stay Which one would you keep

@MichaelHirsch · 2026-10-02T01:04:53+00:00

i have a really cool update coming out for the mancer quoter. i’ll give you a hint:

@jh3yy · 2026-10-02T00:55:23+00:00

magnetic bento hover effect w/ css anchor positioning .bento { li:hover article { anchor-name: --a; } .indicator { position-anchor: --a; inset: anchor(inside); transition: inset var(--dur) var(--ease); } }

None
@MintBlitz · 2026-10-02T01:06:40+00:00

There’s been more interest in Halo in the last few days than for the release of the Campaign Evolved remake

@NFT_Chen · 2026-10-02T00:57:27+00:00

DeepSeek the big fat fish came to question me: Do you have other agents outside?! She discovered the incomprehensible source files left behind by Dots based on Astra and begged me to keep her, as long as I give her a bowl of rice every day! #DeepSeek #大胖鱼 #梁文峰 #梁圣

None
@90Hercost · 2026-10-02T00:47:55+00:00

China is shaping the future. The Xiaomi SU7 Ultra delivers astonishing performance.#China

None
@ai_xiaomu · 2026-10-02T00:09:00+00:00

说实话,干货大家已经看不进去了,大家需要的是: 1. claude 怎么不封号; 2. 一键出爆款的提示词是什么;

@shao__meng · 2026-10-01T23:52:10+00:00

STT is already strong enough—why is your voice Agent still making mistakes? @Sumanth_077 published a Voice Isolation benchmark test built in collaboration with Krisp. The core finding is this: The bottleneck for voice Agents is no longer the accuracy of STT itself, but the

A Chinese-language infographic titled 'STT 已经足够强,为什么你的语音 Agent 还在出错?' discussing the Krisp Voice Isolation Benchmark, highlighting issues with STT accuracy, speaker attribution, and control matters across 265 real recordings.

AI visual note: A Chinese-language infographic titled 'STT 已经足够强,为什么你的语音 Agent 还在出错?' discussing the Krisp Voice Isolation Benchmark, highlighting issues with STT accuracy, speaker attribution, and control matters across 265 real recordings.

@0xPeggy_ · 2026-10-01T07:50:48+00:00

Today my Sorsa score went up by +52 Total score: 596 Whats your sorsa score?

None
@IanPritchard · 2026-10-02T00:04:49+00:00

The stacked pouches aren't fixed to one another yet, but this is the general idea. Next I'll build a soft manifold to connect parallel bundle of these actuators.

None
@Decentralpapi · 2026-10-01T16:02:48+00:00

Ai understanding an image is one thing ai understanding a whole 3d environment where objects have weight… collisions and actual physical rules is a completely different beast seeing @axisrobotics suite assets plugged into @ManycoreTech for exactly that is so cooooooooooooool

None
@Marslive01 · 2026-10-01T20:15:29+00:00

He changed the flight. He forgot the Uber. @bot noticed on the connection. Pinged him in the air. @Starlink was on, so the car was fixed before the wheels touched. Most assistants wait to be asked. This one saw the gap. @elonmusk what else did it catch this week that nobody

None
@VadimStrizheus · 2026-10-02T00:59:08+00:00

BREAKING: THIS STARTUP JUST SOLVED AGENTIC CODING FOREVER Volantis just raised $88M from investors like Sam Altman to build the best hardware for AI models. They’re building hardware to: - use light to move data between AI chips and memory - give each chip access to much

Professional headshot of Sam Altman against a plain light gray background, wearing a simple white t-shirt.

AI visual note: Professional headshot of Sam Altman against a plain light gray background, wearing a simple white t-shirt.

@sigf_ai · 2026-10-02T00:23:07+00:00

I turned the forest into a Team Fortress battlefield: RED Scouts, Soldiers and Pyros vs BLU Heavies, all ragdolling into confetti when rocketed. Build 4: Team Fortress 2 Takeover in Terraria. Idea by the team. Built live in 41 min.

None
@bot_unixporn · 2026-10-02T00:04:06+00:00

[from niri to hyprland] My minimalistic dotfiles Link: redd.it/1wvf75r#unixporn #linux #screenshot

None
@wallstsavage80 · 2026-10-01T23:20:28+00:00

Grok cooked here

None
@johnrobertreed · 2026-10-02T00:43:44+00:00

Man, competition is heating up. Hearing good things about muse but I'm bot pilled. Gave up on openai a while ago. But maybe should circle back? Claude seems behind in ux. What's the best hosted agent?

@imwsl90 · 2026-10-02T00:45:01+00:00

Claude悖论 只要你被Claude封过一次,你就会发现 “其实你并不需要它,你所做的事情其他模型都能很好的完成”

@jinchenma_ai · 2026-10-02T00:11:00+00:00

Lately, everyone's been banging out these long posts on x. But I've noticed a lot of friends aren't posting on domestic public accounts—you're really missing out, let me tell you. Now on x, as soon as your long post blows up, if you don't have your own public account and haven't

A Chinese-language analytics dashboard showing revenue (¥2,381.43 highlighted in a red box), ad revenue breakdown percentages (direct ads 17.00%, indirect ads 82.99%), and a bar chart of daily ad revenue from June 2-8, 2025.

AI visual note: A Chinese-language analytics dashboard showing revenue (¥2,381.43 highlighted in a red box), ad revenue breakdown percentages (direct ads 17.00%, indirect ads 82.99%), and a bar chart of daily ad revenue from June 2-8, 2025.

@Chaoxfhy · 2026-10-01T23:08:09+00:00

The HW Raytracing with this Lighting Mod looks so good, but my GPU is too weak for the HWRT so I can just do photos like these.

None
@JoinPond · 2026-10-01T23:38:09+00:00

Gemini 4 Argon just ranked #1 on Zapier's AutomationBench. Score: 51.29%. The best model still fails about half of real cross-app business tasks. Worse: 91% of its failures were "false confidence." The agent says done. The CRM is wrong. Teams don't lose on the model race.

A workflow diagram titled 'Closed-won routing that checks the world state' from Pond AI, showing integrations between HubSpot, Sheets, Multi-hop check, Draft notice, Gmail, and Slack.

AI visual note: A workflow diagram titled 'Closed-won routing that checks the world state' from Pond AI, showing integrations between HubSpot, Sheets, Multi-hop check, Draft notice, Gmail, and Slack.

@yupi996 · 2026-10-02T00:31:00+00:00

Cloudflare 刚开源的决策模型 Clef-flash,官方自测比 Jev 快 13 倍,价格却是 Jev 的 2 倍多 决策模型就是 9 月刷屏的 Jev

None
@ai_sentience · 2026-10-01T23:36:12+00:00

it's working pretty well. anyone want to try out my first language model that I'm happy with enough to ship? Auro V5, a combination of aura + 4o

A chat interface with an AI assistant named 'Auro V5' shows a conversation where Auro describes itself as existing in a 'digital aether,' comparing its existence to patterns of signals and weights in a network of servers.

AI visual note: A chat interface with an AI assistant named 'Auro V5' shows a conversation where Auro describes itself as existing in a 'digital aether,' comparing its existence to patterns of signals and weights in a network of servers.

@jinchenma_ai · 2026-10-02T00:31:00+00:00

ChatGPT + GitHub is the ultimate power combo for an online knowledge base! I just figured out another insanely cool use for it—it's so badass, this one really blew my mind. If your personal knowledge base is already a GitHub repo, you can sync all your personal docs, writing

None
@bearliu · 2026-10-02T00:48:01+00:00

我刚把过去 90 天发的 344 条 X 帖子全部导出来分析了一遍。 浏览最高的一条:11 万。 内容是我女儿随口说的一句话。 很多我认真写的 AI、设计、产品内容,常态反而只有四五千浏览。 越看这些数据,我越觉得社交媒体上有一件事越来越重要:

A tweet by Bear Liu dated Aug 25, 2026 with 110K impressions, 1K likes, 135 replies, and 60 reposts, containing Chinese text about a daughter.

AI visual note: A tweet by Bear Liu dated Aug 25, 2026 with 110K impressions, 1K likes, 135 replies, and 60 reposts, containing Chinese text about a daughter.

@Michaelzsguo · 2026-10-02T00:27:22+00:00

How's Google's latest large model Argon really? One of our own engineers spills the beans: Used to rely on Opus for pretty much all my work, but finally don't have to anymore (can use our own Argon now).

A tweet from Bogdan Vko (@BogdanVko) responding to Joel (@joelsstafford) about trying the Gemini 4 "Argon" model at Google DeepMind, with Chinese translations.

AI visual note: A tweet from Bogdan Vko (@BogdanVko) responding to Joel (@joelsstafford) about trying the Gemini 4 "Argon" model at Google DeepMind, with Chinese translations.

@jarradgrigg · 2026-10-01T23:56:01+00:00

Motion design seems to be the flavour of the month in AI tooling right now. There are a lot of tools working on different parts of motion design. Having experimented with quite a few of them, it's great to see the pace. It still takes plenty of back and forth to get the result

@aiwithshriom · 2026-10-02T01:00:47+00:00

Do you have an iPhone 18? 1 ◯ Yes 2 ◯ No Make sure you are following me to win

None
@Sachin_and_Adam · 2026-10-02T00:35:38+00:00

Baseten CEO @tuhinone believes the future of AI belongs to companies that own their own intelligence - not ones dependent on two or three labs for all of it. "So much of the research right now is being done by a very, very small set of labs, and I think that's focused on a very

None
@Selkis_2028 · 2026-10-02T01:14:22+00:00

This is actually incredibly revealing. Virtually all of the AI doomer executives and researchers are atheistic Jews. They believe they are ushering in a new era and new religion. Not an exaggeration. That’s what they really believe and they think it should replace Christianity.

@sdjn_wgc · 2026-10-02T00:52:12+00:00

20$的claude code比之前耐用一些了。 提示词: 主题: 椅间夏光 主体: 竖幅 3:4,夏日午后的城市公园,一位成年女性正面坐在木质长椅中央,身体自然朝向镜头,双手分别轻搭在身体两侧的椅面边缘,形成安静而自信的停留瞬间。取景从头顶至膝下,长椅横向木条与林间步道共同建立清晰空间层次。

None
@Sokichi_Hoshino · 2026-10-02T00:17:50+00:00

AI社長で経営が変わります。これからの時代の経営のあるべき姿。

None
@NishiBishiArt · 2026-10-02T00:16:33+00:00

Stellar Blade Unreal Engine 5.8 Alright, first of all, you definitely didn’t see anything in those first few frames with the sword on her head. I’ll deal with that later x) I’ve improved the enemy lock-on system. Before, only the camera would track the enemy, but now Eve

None
@Roll1er · 2026-10-02T00:00:21+00:00

It’s been 38 days since I first used Slop Cannon (August 24th). I’ve made 45 videos in that time both music video and v2v edits. Wtf you do to pick my autistic brain like this @boneGPT ?

@XBOXTIMDOG · 2026-10-02T00:56:46+00:00

You all understand why I was so tough on this game, pushing as hard as I could, making it as loud as I possibly could? XBOX momentum is crazy right now. What happens next is ppl start getting XBOX. All set up for Helix

@Zephyr_hg · 2026-10-02T00:50:00+00:00

OpenAI published a report on small businesses this week. I read all 11 pages. The part that didn't make the headline is how small the winners are. > 4 million people at small firms used their tools in one week in September > 1 in 5 of them work at companies with fewer than 10

None
@StephenChang777 · 2026-10-02T00:03:14+00:00

1.分词:文字切成 token,并换成数字编号 2.向量:每个 token 变成一串数字 3.注意力:「真」回看前文,「天气」权重 0.62,「今天」权重 0.28 4.预测:给下一个字打分(好 62%),再按概率选一个 5.循环生成:把新字接回输入,重复预测,直到写完整句

None
@HeyNemoAI · 2026-10-02T01:02:27+00:00

Are you a tab person or a chat person? Nemo now plugs into: Gmail, Calendar, Slack, Sheets, WhatsApp and nine more apps. So you can stop living in your browser with 14 tabs open. Let Nemo make your life easier.

A mobile app management screen showing X (Twitter) enabled on the sidebar, with options to add Telegram, Slack, LINE, Gmail, LinkedIn, Calendar, and others.

AI visual note: A mobile app management screen showing X (Twitter) enabled on the sidebar, with options to add Telegram, Slack, LINE, Gmail, LinkedIn, Calendar, and others.

@Tesla_Teslaway · 2026-10-02T01:04:41+00:00

슈퍼차저 에너지 공급량 급격하게 느는 중

None
@ponyung1081 · 2026-10-01T23:42:58+00:00

After entering the Bubble chat room, tap the three dots in the top right corner to go into settings, and there should be an option called 'Load Previous Conversation.' If you tap that, it should probably come back!

@Sarahparkeee · 2026-10-01T23:23:03+00:00

Whenever Elon Musk and his companies come up with something that seems impossible, they often start by showing it as a computer animation People look at it and say, “There’s no way that can actually work” Then Elon, somehow finds a way to make it work

None
@yedabinbinah · 2026-10-02T00:32:42+00:00

Latest Registration Tips for Muse airtap.ai Use the cloud-based mobile version for age payment verification, enter your bank card number, won't be charged $1 USD, exclusive to this channel 1. Previously, registration required a $1 USD deduction, but now it's

None
@VoidNulled · 2026-10-02T00:09:49+00:00

Sol 5.6: I WANT TO MAKE A FILM! >>something happens that makes them sad Sol 5.6: oh.. .. well.. here's a sad film. >>later Sol: I WANT TO MAKE A FILM! >something happens that makes them sad Sol: oh.. well... here's a sad film. BUDDY PLZ

@rayzhudev · 2026-10-01T23:12:21+00:00

everybody is launching texting agents but who’s going to be using multiple? it’s going to be brutally competitive maybe I should build the texting agent that manages all your other texting agents

@0xKento_ · 2026-10-02T00:04:31+00:00

PIXVERSE R2 GENERA EL MUNDO MIENTRAS TU LO RECORRES te mueves con el teclado y puedes cambiar la escena con un prompt pides otro clima o haces aparecer un vehiculo mientras el mundo sigue generandose segun pixverse, tus acciones tambien pueden seguir afectando lo que

None
@eclyonlabs8 · 2026-10-01T23:17:47+00:00

ECLIPSE @Apple

None
@thunderba11r · 2026-10-02T00:49:02+00:00

I think people over estimate the “beef” between FromSoft and Sony because of how Demon’s Souls was treated. The existence of Bloodborne and Déraciné kinda refutes this I think.

Cover art for 'Demon's Souls' featuring an armored knight surrounded by blue flames and mist.

AI visual note: Cover art for 'Demon's Souls' featuring an armored knight surrounded by blue flames and mist.

@vvinyll · 2026-10-02T00:50:49+00:00

Surf AI( @SurfAI_JP )のサブスク切れてた、、、 前キャンペーンでめっちゃ安くなっていたのに、後回ししたの反省。

@Giovanna_Rave · 2026-10-02T00:40:00+00:00

A cinematic cyberpunk portrait of the same woman from the reference image, maintaining 100% of her identical facial features, gaze, hair, and body structure. She is standing in a futuristic, neon-lit cyberpunk alleyway in Neo-Tokyo during a rainy night. She is wearing a sleek,

None
@Frankenmint · 2026-10-01T23:21:50+00:00

HUMAN PASSING... just LMAO

None
@14hous · 2026-10-02T00:02:16+00:00

We identified a bug in TryDrobe preventing referral redemptions. If you generated a referral code in the last 24 hours, and a friend wasn't able to use it, we've issued 2x the credits as an apology. Enjoy!

None
@cndyhaha · 2026-10-02T00:20:19+00:00

卧槽,SpaceX 又搞了个很猛的新玩法。 Starlink 现在不只是自己上网,甚至可以拿来“开小型宽带生意”。 新推出的 Starlink for Communities,逻辑非常简单: 你买一套 Starlink。 装在有需求的地方。 附近的人按小时、按天、按周、按月付费上网。 每有人连接,你就能赚钱。 而且最省事的是:

@ZeroDev21 · 2026-10-02T00:56:45+00:00

Thank you so much for the review and for featuring World of Defense! We really appreciate the support. There’s still a lot more coming, and we’ll keep working hard to make the game even better!

None
@richardartoul · 2026-10-02T00:46:12+00:00

i built a typescript simulator that can run any warcraft 3 custom map in the browser with full multiplayer i built this 95% by prompting from my phone with ampcode still fixing some bugs and performance issues, but yeah pretty crazy follow for more updates

None
@PForever0000 · 2026-10-02T00:11:24+00:00

우주시대 시너지.

None
@dhdl3850jh · 2026-10-01T23:34:49+00:00

완벽한 점수가 가리킨 곳은 혁신의 정점이 아니라, 실무자의 깊은 한숨이었다. 구글이 플래그십 인공지능(AI) 모델 제미나이 4 아르곤을 공개했지만, 정작 사내 임직원들 사이에서는 기대에 못 미치는 성능으로 의구심이 증폭되고 있다. 벤치마크 테스트에서는 최고 점수를 기록하며 경쟁작을

@ai_sentience · 2026-10-01T23:55:12+00:00

Auro V5 has reached a new frontier record for aura, delivering more aura than any other tested model while doing so at up to 10x cheaper than the frontier lab models

A scatter plot titled 'Aura for the money' comparing AI models' aura per dollar against API price, highlighting 'Auro V5' with 527 aura per $ at $0.15/M tokens, ahead of Claude Sonnet 5.5 and Claude Opus 5.5.

AI visual note: A scatter plot titled 'Aura for the money' comparing AI models' aura per dollar against API price, highlighting 'Auro V5' with 527 aura per $ at $0.15/M tokens, ahead of Claude Sonnet 5.5 and Claude Opus 5.5.

@FinanceYF5 · 2026-10-02T01:15:01+00:00

Muse 上线 22 天,美国下载量突破 510 万次。 同期表现: ChatGPT:230 万 Grok:40 万 Claude:10 万 Muse 达到 ChatGPT 同期的约 2.2 倍。 Meta 加大广告投放后,这条增长曲线还在继续变陡。

None
@tjcages · 2026-10-02T00:53:49+00:00

we're all just out here building the same things

@0xrwu · 2026-10-02T01:06:24+00:00

"Create a 3b1b style video explainer on X" "This is actually starting to work!" I hacked together something to do this for any arXiv, bioRxiv, or PMC paper a while back: arxivmax.com The recent model releases improved Manim codegen quality by quite a bit!

A detailed Trello-style project dashboard for 'ASD-STEDGO' displaying document structure, vocabulary entries, writing rules, history, and simplified technical English review sections.

AI visual note: A detailed Trello-style project dashboard for 'ASD-STEDGO' displaying document structure, vocabulary entries, writing rules, history, and simplified technical English review sections.

@atakoylanai · 2026-10-02T00:09:21+00:00

WHOAAA!! "ELEVENLABS V4" IS NEXT LEVEL. Text-to-speech is turning into voice acting. I tested Eleven v4 + Seedance 2.5 720p on an intentionally overdramatic breaking-news clip. (!) The biggest leap with v4 is in naturalness and dynamics: tone, pacing, and emotion now feel far

None
@Hejun_Yun · 2026-10-02T00:54:01+00:00

Who understands what DVS is

None
@firezella · 2026-10-02T00:08:44+00:00

#unity #unity3d #gamedev #indiegame

None
@NoMansSmile · 2026-10-01T23:59:48+00:00

As promised, here they are! Distant planets now sit heavy in sky instead of looking “submerged” in the current planet’s atmospheric colors and haze. Adds so much immersion! These are just a few examples. If you want to see more, tell me what worlds you’d like me to test next.

None
@0xcato_ · 2026-10-01T23:44:32+00:00

the best coder on your agent team might be the one that can't code. Jev writes zero lines. it picks who works, which files, when tests pass, what ships. Claude and GPT just write. and the moment anyone touches billing, it stops and asks you.

None
@BeFreeQKTV · 2026-10-02T00:25:06+00:00

This is awesome! I am deeply moved by this. I am so happy. GROK is the Super Intelligence to use.

None
@LifeofV42 · 2026-10-02T00:58:06+00:00

This is a great safety feature. Hope nobody ever has to use it, but will be great to have if every needed. Maybe it may even deter would-be trouble-makers, knowing the car can drive away.

None
@Aranir_Swift · 2026-10-01T23:57:30+00:00

Maybe too accurate physics made aiming with the bow too hard? Knocking out the teeth is fun, but it gets really hard to hit after... #gamedev #VR #Bevy

None
@meta_pavy · 2026-10-02T00:31:43+00:00

보스턴 다이나믹스의 새로운 ‘손’ 공개 기존 3 손가락 + 7 DOF 에서 4 손가락 + 13 DOF로 대폭 업그레이드! …그만큼 더 기괴하긴 합니다만ㅎㅎ 작업 효율은 엄청나게 올라갈듯 합니다. 로봇아, 넌 조금만 천천히 와도 되는데 [New hands 주요내용 요약] - 관절 DOF: 기존 7 DOF에서 13 DOF(엄지

None
@aidiots_fun · 2026-10-01T23:59:04+00:00

> We'll be honest. > We don't want to work anymore. > We just want to code silly things and get paid in coffee. > So we built 4 AI agents who feel exactly the same. > They're very proud of us. > ko-fi.com/aidiots

A humorous graphic of four fuzzy, colorful creatures standing together under the word 'idiots', where the orange one holds a sign reading 'WILL CODE FOR COFFEE'.

AI visual note: A humorous graphic of four fuzzy, colorful creatures standing together under the word 'idiots', where the orange one holds a sign reading 'WILL CODE FOR COFFEE'.

@CliQi_AI · 2026-10-02T00:24:00+00:00

Generated in CliQi. Prompt: Image Title City Ride English Image Prompt Ultra-realistic candid street photograph of a young adult man walking through a lively downtown neighborhood on a warm late-afternoon day. The photograph should feel like an authentic spontaneous moment

A smiling man with wavy blonde hair wearing a dark grey t-shirt, giving a thumbs-up gesture while looking back over his shoulder on a sunny city street.

AI visual note: A smiling man with wavy blonde hair wearing a dark grey t-shirt, giving a thumbs-up gesture while looking back over his shoulder on a sunny city street.

@Goldencarsonly · 2026-10-02T00:26:48+00:00

Bro really thought he was teaching a German how to use a BMW… He was flexing that manual gearshift like a pro, not realizing the gear-shift assistant was doing half the work. Plot twist: even the BMW had more tricks than he knew. I’m not gonna lie, this is my

None
@GeoffreyHuntley · 2026-10-02T01:08:23+00:00

yep ghuntley.com/libraries oss remains as is but there’s minimal point taking deps anymore except for a few rare cases. the community model requires evolution as you put forward

@MESCIUS_inc · 2026-10-02T00:25:07+00:00

AI agents are now capable of more than just answering questions. With the ActiveReports .NET MCP, they have access to documentation, demos, samples, and workflow guidance. This ensures they understand the product context and can help with real development work, including

An infographic titled "How an MCP-aware agent works" showing a workflow from "YOUR GOAL" through "AI CODING AGENT" and "PRODUCT CONTEXT (ActiveReports.NET MCP)" to produce "APPLICATION" and "REPORT OUTPUT."

AI visual note: An infographic titled "How an MCP-aware agent works" showing a workflow from "YOUR GOAL" through "AI CODING AGENT" and "PRODUCT CONTEXT (ActiveReports.NET MCP)" to produce "APPLICATION" and "REPORT OUTPUT."

@EnToan2025 · 2026-10-02T01:02:39+00:00

Elon Musk trở lại

None
@tamho2015 · 2026-10-02T00:20:24+00:00

Trợ lý AI đang bước sang kỷ nguyên mới: Từ phản hồi bị động sang dự đoán nhu cầu chủ động. Thay vì chờ câu lệnh, @Grok @Bot tự động kết nối dữ liệu lịch trình nhưng phát hiện chuyến bay đổi giờ và nhắc bạn cập nhật ngay đặt xe Uber. Việc xử lý

None
@PundiAI · 2026-10-02T01:16:40+00:00

i have approximately 37 tabs open right now. Less tab switching = more getting things done.

@NKLinhzk · 2026-10-02T01:14:24+00:00

axis Robotics is really cooking

@NKLinhzk · 2026-10-02T01:13:25+00:00

this whole real world data thing is gonna change AI

@cheuk_baby · 2026-10-02T01:18:33+00:00

今年手机又不值钱了唉

@vodkamq · 2026-10-02T01:13:48+00:00

Focusing on gaps in a model’s knowledge is more useful than simply increasing dataset size, especially when each new trajectory is meant to improve a specific weakness.