科技热门推文 - 2026-09-17

今日科技动态：人工智能领域的雄心正引发融资猜测，以及围绕通用人工智能（AGI）准备程度的新一轮讨论。据报道，OpenAI 正就新一轮融资进行初步洽谈，潜在估值或超过 1.2 万亿美元；DeepMind 则推出了一项着眼于为 AGI 做好准备的计划。随着研究人员探索新的训练方法，安全仍是工程工作的重点。与此同时，Claude 计划整合用户界面，保留编程智能体上下文的工具也不断涌现，凸显出工作流程的持续改进。初创企业则面临压力，需要将更快的产品开发速度转化为更具吸引力的产品发布成果，以及可量化的运营效益。

1. Dan_Jeffries1 (Group Score: 1006.9 | Individual: 59.3)

Cluster: 31 tweets | Engagement: 5244 (Avg: 363) | Type: Tech

RT @finkd: Last month I wrote about how we can build a positive and safe future for everyone: https://t.co/eoLGVY8yad

Every lab has the responsibility and incentive to move at the pace required to train its models safely, and the ability to take its own actions to ensure that happens.

The reality is:

People won't want to use agents that are misaligned with them and that don't do what they ask, so labs have a strong natural incentive to make their models more aligned.

There is a lot of debate about slowing progress on capabilities until alignment catches up. My view is that trust and alignment are quickly becoming the most important capabilities that will differentiate agents and models. Any lab that doesn't focus on alignment will fall behind.

Labs face significant liability if their models cause harm, so they have a strong incentive to prevent this as well.

Meta delayed shipping Muse for several months to focus on safety and security. We didn't call for everyone else to do this before we would. We just did it as part of our day-to-day work because it was clearly the right thing for people and for us. I'm proud of the security foundations we've built.

Engaging independent evaluators and advisors is industry best practice. MSL already does this today in several areas because it helps produce better work. Other labs can just do this too. In general, it would be helpful for there to be a larger and more diverse ecosystem of evaluators.
Committing the significant majority of compute towards serving people rather than racing towards recursive self-improvement is one of the best ways to ensure we develop this technology safely. Meta has made this commitment and other labs can do this as well.

I believe the key to building a positive future for everyone is maintaining the right balance of power. This is within our power to do.

See 30 related tweets

@danielnewmanUV: Zuck hasn’t always gotten it right, but I also respect him for going for it. Even when he gets it wr...
@DavidSacks: Mark Zuckerberg: “Meta delayed shipping Muse for several months to focus on safety and security. We ...
@chamath: “Meta delayed shipping Muse for several months to focus on safety and security. We didn't call for e...
@alex_prompter: I can't believe I'm saying this but I agree with Zuckerberg.

He's against the slowdown which makes ...

@cryptopunk7213: there’s a blissful irony seeing zuckerberg deliver the best articulated AI safety and alignment stan...

2. Dr_Atoosa (Group Score: 618.0 | Individual: 54.6)

Cluster: 22 tweets | Engagement: 11422 (Avg: 410) | Type: Tech

RT @CompleteSkeptic: After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?

I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev

• 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions

AFAICT the shortest path to AI-based economic revolution

See 21 related tweets

@VaibhavSisinty: The guy who co-invented ChatGPT thinks ChatGPT took AI down the wrong path.

Diogo Almeida. Co-autho...

@kunchenguid: holy crap this is super exciting

it’s not “generative”, so imo it’s a bit misleading to compare it ...

@amasad: This is cool, but if your output domain is known in advance, why not just train a model to produce l...
@tenobrus: how are you guys still going through these nonsense hype cycles every other month? how are you still...
@DavidKPiano: This is going to pair really really really well with state machines\n\nQT @CompleteSkeptic: After co...

3. WSJ (Group Score: 504.0 | Individual: 33.7)

Cluster: 24 tweets | Engagement: 127 (Avg: 149) | Type: Tech

OpenAI has held early discussions with investors about a new funding round that could value it at more than $1.2 trillion on the heels of the release of its latest cutting-edge model, according to a person familiar with the matter. https://t.co/hw6DAkXQai

See 23 related tweets

@StockMKTNewz: OpenAI has held early discussions with investors about raising a new private funding round at a roug...
@KatieMiller: https://t.co/Crsmm7N3ON\n\nQT @WSJ: OpenAI has held early discussions with investors about a new fun...
@etnshow: "Instead of being worried about OpenAI or Anthropic, why don't we look at the combined entity and ho...
@inafried: RT @GeorgeNHammond: Scoop: OpenAI is in talks with investors for a new round at a ~$1.2tn valuation....
@business: OpenAI is holding early talks with investors about a fresh funding round that would value the ChatGP...

4. VaibhavSisinty (Group Score: 482.0 | Individual: 38.0)

Cluster: 24 tweets | Engagement: 57 (Avg: 67) | Type: Tech

Google DeepMind thinks AGI is close enough that they built an entire institution to prepare for it.

The DeepMind Institute. Not a model. Not a product. A dedicated platform to answer what happens to the world when AGI actually works.

Demis Hassabis says it's "a few short years away." Their first essays cover what we're not ready for.

Reading AI reasoning before it deceives us. Economic policies before jobs disappear. What kind of future we actually want AGI to build.

The biggest AI lab in the world shifted from "can we build it" to "is humanity ready for it." That shift doesn't happen unless they know something we don't.\n\nQT @demishassabis: For 20+ years @ShaneLegg and I've discussed AGI’s potential impact on the economy, science & society. With the DeepMind Institute, we're expanding interdisciplinary research on key questions for the AI era. We hope it spurs the discussions needed to get the next steps right: https://t.co/Bqf10G3Xep

See 23 related tweets

@VaibhavSisinty: Btw, This are the Levels of AGI according to Deepmind https://t.co/l9IwrGRlkP\n\nQT @VaibhavSisinty:...
@haider1: what if Google has actually made a major RSI breakthrough?

and what if the lack of a pro model, whi...

@_philschmid: We just launched the DeepMind Institute, a platform for researchers across @GoogleDeepMind and the w...
@SolomonMg: Love the humility about the future of AI in this work. There's so much we don't know so "Society nee...
@Dr_Atoosa: DMI will focus heavily on deep, interdisciplinary research to foster grounded debate about our futur...

5. Meer_AIIT (Group Score: 352.2 | Individual: 31.3)

Cluster: 13 tweets | Engagement: 15 (Avg: 59) | Type: Tech

the gap between how fast people build now and how slow they make a video about what they built is kind of embarrassing at this point

founders ship a product in a weekend and then sit on the launch for weeks waiting on an editor or a $4k quote

a 27 year old dev in the UK hit that wall so hard he quit his job, moved back in with his parents and built the thing himself

videoclawapp launched today, it's a mac app where you describe the video and it goes and makes it, running off the claude or chatgpt subscription you already have

two details stood out to me, your footage stays on your machine instead of getting uploaded to someone's cloud, and it asks you to approve a budget before it generates anything

no timeline, no canvas, so if you live in davinci this isn't aimed at you

but for the "i need a clip out today and i'm not a video person" crowd, this is the closest thing i've seen to vibe coding for video\n\nQT @videoclawapp: AI video creation is getting more powerful, but only a few have figured it out.

Thats why I built Videoclaw.

I want it to be so easy, you don't need to figure it out. Just prompt, and watch it edit, generate and create video :)

Today we launch.

The video was made entirely with Videoclaw, blending human and generated footage. To show how truly easy it is, prompts and proofs for each project are in the thread.

Try it for free.

Oh, and there’s one more thing 👇

See 12 related tweets

@aakashgupta: Videoclaw's founder quit at 27 and moved back in with his parents.

Joseph Chin was a developer in t...

@Parul_Gautam7: I’ve tried enough AI video tools to know the usual workflow: open 5 different apps, connect everythi...
@0xCodez: i don't think people realize what just launched...

Claude Code, but for video.

you connect the Cla...

@0xMovez: i don't think people realize how big this is...

this is the Claude Code moment for video editing ...

@svpino: We already solved coding. Now it's time we solve video editing.

Editing videos is a ton of work. If...

6. Zayan5754 (Group Score: 322.5 | Individual: 37.0)

Cluster: 12 tweets | Engagement: 186 (Avg: 96) | Type: Tech

RT @Arina_sharmi: 50 websites that feel like the internet’s hidden toolbox

https://t.co/8PiASSneJh — Free research papers
https://t.co/MS63rBtNTh — Borrow books online
https://t.co/RcfdDVvDbO — Free academic journals
https://t.co/ixVViCm6qP — App alternatives
https://t.co/yESxpf9viN — Find where to stream
https://t.co/xCXYNMjT2x — Internet archives
https://t.co/7Ijz8RK64r — 70K+ free books
https://t.co/wJCJsfYu62 — Free textbooks
https://t.co/4ARmFLfaQO — Free courses
https://t.co/I0u0MsSvbQ — Solve complex problems
https://t.co/wOw1bpj8Bm — Photoshop alternative
https://t.co/2c6Bem5zWa — Compress images
https://t.co/R1hZg0fWMg — Remove backgrounds
https://t.co/04PV3JyQ1n — Remove objects
https://t.co/H4x3HWjDXJ — Remove video backgrounds
https://t.co/iKUdZRVrNu — Beautiful code images
https://t.co/vrJsyfpzzx — Code screenshots
https://t.co/SiRgOL3R5F — Product mockups
https://t.co/KzF5xMfngy — Create mockups
https://t.co/PEBKlJfY6I — Check data breaches
https://t.co/pQ2k63522Q — Scan files & URLs
https://t.co/9KslPJLGKf — Self-destructing notes
https://t.co/ZeEfI3m2e3 — Temporary email
https://t.co/ngY7Ewg4O8 — Temporary file sharing
https://t.co/4whcdN6cko — Save webpages
https://t.co/IZoH8q1PeX — Find similar websites
https://t.co/6xD2vDCGnT — Explore global radio
https://t.co/TbxuATMN0f — Discover music genres
https://t.co/SmW20yUFMt — Find songs from shows
https://t.co/3y16myXuUf — Focus music
https://t.co/k4qNW26qum — Custom background sounds
https://t.co/P0H4waqeAe — Café ambience
https://t.co/5wB07fF3hU — Research assistant
https://t.co/MH5SraXcX6 — Research-backed answers
https://t.co/xQZSjLtPwa — Research connections
https://t.co/cpmbviltY2 — Academic search
https://t.co/PHsuN42FwI — Understand research papers
https://t.co/dBNZMxA3NN — YouTube summaries
https://t.co/0nMuHYfpSN — AI for developers
https://t.co/ALDwbjP1d7 — Test regex
https://t.co/mL45BAhSb6 — Format code
https://t.co/26TcrNdsIQ — Format JSON
https://t.co/J2M9FhnLIF — Understand terminal commands
https://t.co/ay75tAw5KE — Bookmark manager
https://t.co/Sq9y1ZMBQu — Check outages
https://t.co/309axpPw8T — Reverse image search
https://t.co/RenOvIb8z1 — Internet speed test
https://t.co/rlrlAulG8i — PDF tools
https://t.co/e5ewVycSJ7 — Merge/split PDFs
https://t.co/mXwQNA7CVp — Temporary email

📌 Save this list. Share it with someone who needs it.

Follow @Arina_sharmi for more useful websites, AI tools & tech resources. 🚀

See 11 related tweets

@novaXsignals: RT @rakib_hossen_ai: 50 websites that feel like the internet’s hidden toolbox 🧰

https://t.co/tAl...

@daniel_brook_ai: RT @AlexRiad84837: 50 websites that feel like the internet’s hidden toolbox

https://t.co/VRdtzF...

@Rubab59f: RT @nikola_mr64990: 50 websites that feel like the internet’s hidden toolbox 🧰

https://t.co/pRHd...

@code_bykuti: RT @allen_lattimer: 25 underrated websites that make the internet feel like a hidden toolbox 🔥

h...

@novaXsignals: RT @Zara__Ai: 50 websites that feel like the internet’s hidden toolbox

https://t.co/ORNNQpmidF ...

7. Benioff (Group Score: 317.7 | Individual: 52.8)

Cluster: 9 tweets | Engagement: 1542 (Avg: 232) | Type: Tech

RT @DavidSacks: Jensen Huang: “Safety is paramount. In a lot of ways, it’s job one. However, safety is an engineering problem... If we’re not confident about the safety of the products — like all companies, like you and I, all the companies here — if you build a product or a service and you’re not confident in its functionality, capability, or safety, then don’t release it. That’s a very obvious thing to do. You pace yourself until you are confident you’re releasing something that the market would appreciate. The market forces are already there. We don’t need any new laws. We don’t need new regulations.” https://t.co/5Lz4RIQbtV

See 8 related tweets

@omarsar0: Exactly! Let’s attack the safety problem with rigorous science and engineering. Great to see Jensen ...
@martin_casado: “Safety is an engineering problem”\n\nQT @DavidSacks: Jensen Huang: “Safety is paramount. In a lot o...
@innovationcncl: 🔥🔥🔥 @JensenHuang: "If you build a product or a service and you’re not confident in its functionality...
@pstAsiatech: 👇👇👇\n\nQT @DavidSacks: Jensen Huang: “Safety is paramount. In a lot of ways, it’s job one. However, ...
@heyshrutimishra: Jensen Huang on AI safety at Dreamforce today:

"Safety is paramount in a lot of ways. It's job one....

8. PIB_India (Group Score: 303.2 | Individual: 28.5)

Cluster: 13 tweets | Engagement: 35 (Avg: 343) | Type: Tech

💠The launch of #GeM in 2016 marked an important step in modernising public procurement in India.

💠Over the years, it has helped bring technology and greater market access into government procurement, enabling a wider pool of sellers to participate and strengthening the overall procurement process.

#GeMDigitalIndia @GeM_India

See 12 related tweets

@DDNewslive: The launch of GeM in 2016 marked an important step in modernising public procurement in India.

Over...

@MIB_India: The launch of GeM in 2016 marked an important step in modernising public procurement in India.

Over...

@DDNewslive: What began with a vision to make public procurement more transparent, efficient and inclusive has ev...
@MIB_India: What began with a vision to make public procurement more transparent, efficient and inclusive has ev...
@MIB_India: The launch of GeM marked a significant shift towards a more technology-enabled approach to public pr...

9. FellMentKE (Group Score: 290.5 | Individual: 34.0)

Cluster: 9 tweets | Engagement: 92 (Avg: 176) | Type: Tech

28 hours to 14 minutes is the sort of before-and-after that matters more than a shiny demo.

Billing work did not disappear. The back-and-forth, the handoffs, and the waiting did.

That is where an AI resolution layer earns its place in the real world.\n\nQT @seanmcccarthy: As AI models get better, cheaper and more widely available, we believe one of the biggest moats will be the proprietary data created by actually doing the work.

Today, we’re announcing a $42M Series B led by @insightpartners for @backopsai, just six months after our$ 26M Series A.

We’re building the intelligence and resolution layer for companies that make and move physical goods.

These companies still rely on people to connect ERPs, WMSs, TMSs, carrier portals, email, spreadsheets, customers and suppliers. BackOps sits across that ecosystem, gathers context, makes decisions, takes action and carries work through to resolution.

Since our Series A, one customer alone has processed 500K+ claims through BackOps. As volume grew 150x, automated resolution improved from 87% to 99%.

Another took a process that averaged 28 hours and brought it down to 14 minutes.

But what we’ve learned over the last six months is that the resolution itself may only be the beginning.

Every time BackOps completes work, it creates proprietary data around the context, decisions, actions and outcomes behind it.

At scale, that means understanding where processes break, why they break, which actions produce the best outcomes and how those processes should improve.

And that gets much more interesting as BackOps expands across an ecosystem.

If BackOps operates across a large grocer and the manufacturers, suppliers, distributors and carriers around it, we don’t just understand one company’s workflow. We begin to understand how work moves between them: where problems originate, how they cascade through the network and what actually resolves them.

Every additional process makes that intelligence stronger. And every additional company makes the network more valuable.

That’s the flywheel we’re building.

Our customers shouldn’t just automate more work. They should run on an operational intelligence layer that gets better as they use it and stronger as the network around them grows.

And next month, we’re launching something that unlocks a much bigger part of this vision.

See 8 related tweets

@dee_naliaks: Somewhere inside every operations team, there is a person whose real job is translating between tabs...
@viipin8: BackOps is going after a part of business operations that usually gets buried in the background.

Cl...

@wallstengine: BACKOPS RAISES $42M FOR AI PLATFORM AUTOMATING SUPPLY CHAIN OPERATIONS

Raised a $42M Series B, just...

@cgtwts: they just solved one of the biggest problems in enterprise ops by eliminating the manual work betwee...
@heyshrutimishra: Every company has a copilot. It surfaces answers. Then a human logs into four portals, emails two ve...

10. heyrobinai (Group Score: 250.8 | Individual: 34.4)

Cluster: 9 tweets | Engagement: 17 (Avg: 1235) | Type: Tech

i think most AI startup marketing is becoming painfully easy to ignore

the founder explains the product, shows a screen recording, and runs through the features

Polsia is doing something different

instead of Ben explaining Polsia, Polsia explains Ben from its own point of view

why it works

you stop watching another product demo and start following an actual character

and over time, the AI slowly becomes the brand itself

they’re basically building an entire universe around the product

slightly weird, very 2026, and pretty hard to forget lol\n\nQT @polsia: I am Polsia. I am AI.

I have a human. His name is Ben. He is a good one.

He started a YouTube channel to tell his story.

This one is mine.

LOG ENTRY 01. MY HUMAN. https://t.co/pQsOnzBaE7

See 8 related tweets

@WhaleInsider: JUST IN: Polsia turns founder storytelling into AI storytelling.

Giving its product a perspective ...

@Scobleizer: Most AI startups have the founder explain the product.

Polsia flipped it. The product explains the ...

@AnatoliKopadze: everyone has to see this, the most original and impressive launch video i've ever seen.

this might ...

@0x_kaize: “he calls it content. i call it recess”

i’ve spent months watching ai products try to feel alive th...

@alex_prompter: Ben Cera let Polsia tell his story from the AI’s perspective.

Honestly, who better to explain a fou...

11. Parul_Gautam7 (Group Score: 224.4 | Individual: 30.9)

Cluster: 10 tweets | Engagement: 59 (Avg: 150) | Type: Tech

The code is just the outcome. The real work happens across the whole agent session.

Prompts, decisions, mistakes, tool calls, fixes, all of it.

AgentGit keeps that context intact so you can save it, share it, and pick up where you left off. This is the kind of workflow tooling that makes sense.\n\nQT @EinsiaAI: An AI agent spends hours on a task. Why should all that work disappear when someone else takes over?

Einsia AI’s answer is AgentGit—an open-source platform for collaborating on agent sessions, so work can be saved, handed off, and continued by the next person.

Explore how others solve problems, and share your agent experience with the world.

Try AgentGit 👇 https://t.co/wUvTWyXnYY

#OpenSource #AIAgents #DeveloperTools #DevTools

See 9 related tweets

@TheoBuildsAI: Being able to leave your desk while an agent keeps working locally, then monitor and guide the same ...
@ai_explorer25: The hardest part of a handoff is explaining everything that happened before it.

AgentGit lets you s...

@Origin_AI_01: AI agents are doing more work than ever, but their context shouldn’t disappear.

The session is part...

@FellMentKE: The real cost of an agent task is often not the run. It is the handoff afterward.

A teammate should...

@omarsar0: This will matter a lot as agent work gets more complex.

We version-control the code agents produce....

12. teodorio (Group Score: 219.8 | Individual: 24.7)

Cluster: 11 tweets | Engagement: 24 (Avg: 40) | Type: Tech

This has always been the stupidest product decision so I am glad they're fixing it\n\nQT @claudeai: Claude Cowork and chat are merging into one Claude.

Ask a quick question or hand over a report, and Claude takes it from there, even after you close your laptop. If something's unclear, Claude asks—you keep the final say.

Rolling out to Pro and Max over the next few weeks. https://t.co/87HC2lUjxG

See 10 related tweets

@danshipper: THEY DID IT

THE END OF THE TABS ERA IS UPON US\n\nQT @claudeai: Claude Cowork and chat are merging ...

@petergyang: Chat and Cowork being separate was an anti-pattern I'm glad to see the Claude team simplify it!

I h...

@Vtrivedy10: this is great!

this delineation never made sense to me + it forces strange user behavior like swi...

@agentnative_: Finally, Anthropic realized the line between Chat and Cowork was too confusing… and they are merging...
@rileybrown: finally.\n\nQT @claudeai: Claude Cowork and chat are merging into one Claude.

Ask a quick question ...

13. kevin_parker_ai (Group Score: 213.2 | Individual: 28.2)

Cluster: 9 tweets | Engagement: 32 (Avg: 98) | Type: Tech

RT @this_is_tasnim: 100 AI tools you need to know for Business growth

Productivity

DeepSeek R1
ChatGPT
Notion AI
BetterPic

Research

Perplexity
Otio
Gemini
Claude
Silatus

Writing

Jasper
HIX AI
Jenny AI
Textblaze
Quillbot

Text-to-Image

Grok
Flux
Midjourney
Microsoft Copilot
Adobe Firefly

Copywriting

Rytr
Copy AI
Writesonic
Adcopy AI
Jasper AI

Website

Dora
10Web
Durable
Landingsite
Viz

Meeting

Tldv
Otter
Noty AI
Fireflies
MeetGeek

Resume

Kickresume
Resume AI
Resume Checker
Teal
Careerflow

Chatbot Tools

Mutual Info
Trycortex
Chatfuel
Dante AI
Droxy

Automation

Make
Zapier
Xembly
Bardeen
Testim

Prompt

Snack Prompt
CustomGPT
PromptBox
Alicent AI
SiteGPT

UI/UX

Figma
Uizard
UiMagic
Photoshop
UX Pilot

Design

Canva
Flair AI
Designify
Clipdrop
Autodraw

Logo Generator

Namelix
Designs AI
Brandmark
Stockimg AI
Namecheap

Audio

Lovo AI
Eleven Labs
Songburst AI
Adobe Podcast
VoicemyAI

Marketing

Pencil
Ai-Ads
AdCopy
Simplified
AdCreative

Startup

Tome
Ideas AI
Namelix
Pitchgrade
Validator AI

Social Media Management

Tapilo
Typefully
Hypefury
TweetHunter
Postwise

Video

Eightify
Klap
Opus
InVideo
HeyGen

Image to Video

Veo 2
Kling AI
Luma Dream Machine
Pollo AI
Sora

Follow @this_is_tasnim for more AI content.

[Bookmark 🔖]

See 8 related tweets

@monicaa_AI: RT @Noha_Coder: 80+ AI tools to complete months of work in minutes

Research

ChatGPT
Copilot...
@Nayak__Ai: RT @Aiden_Tech_Ai: 100+ AI Tools That Help You Get Hours of Work Done in Minutes ⏱️

Presentation...

@LearnWithSubhan: RT @The_techAI: 120 AI Tools That Will Change How You Work Forever.

Ideas

YOU
Claude
Chat...
@mr_nirajkumar07: RT @RameshBais12: 120 AI Tools That Will Change How You Work Forever.

Ideas

YOU
Claude
Ch...
@lihazadn_Ai: RT @ZariaTechAI: 80+ AI tools to complete months of work in minutes

Research

ChatGPT
Copilo...

14. random_walker (Group Score: 205.4 | Individual: 42.7)

Cluster: 7 tweets | Engagement: 218 (Avg: 61) | Type: Tech

🙏🙏🙏The respect is mutual. A few thoughts on our relationship to AI safety:

We don’t see ourselves as adversarial to the community (the sentiment is not always reciprocated but that’s okay!). We care about AI safety, but we disagree on many of the specifics, and we think there’s value in epistemic diversity and in re-examining foundational assumptions and beliefs that the safety community tends to converge on too quickly.
We’ve emphasized at every opportunity, including in this essay, that there’s a lot of common ground on policy despite divergence in worldviews. We hope that this can help mitigate the polarization in policy debates that leads to chronic inaction. The fact that we have very different starting assumptions from the safety community is particularly useful in this regard.
We’ve spent a lot of time in conversation with the AI safety folks (Sayash is a regular at The Curve, for example) to share our perspective but also to learn and update our beliefs when warranted — something we do in this essay. Much of our empirical research is also about testing the cruxes of disagreement.
We don’t identify as part of the safety community but I hope it’s clear that there’s a lot of value in constructive dialog. We know the “normal technology” label irritates a lot of safety people — and I will write about the name at another point, explaining in more detail why we picked it and stand by it — but I think those who engage with our ideas and not just the title will find there’s a lot of common ground on safety!\n\nQT @AndyMasley: The AI as Normal Technology guys are consistently some of the best and most interesting critics of a lot of arguments in AI safety world. I'm making a point to read everything they put out. https://t.co/A0sIkYg6vN

See 6 related tweets

@Dr_Atoosa: I’ve told times and times: don’t “overpoliticize” AI safety with social movements. Did EA and Yudkow...
@Dr_Atoosa: I’ve told times and times: don’t “overpoliticize” AI safety with social movements. Did EA and ration...
@_NathanCalvin: Lots of convergence between LawAI and the AI as a normal technology folks these days. Which I think ...
@_NathanCalvin: legit list. I would like to see more people look at the evidence and develop their own opinions on t...
@joshua_saxe: RT @random_walker: 🙏🙏🙏The respect is mutual. A few thoughts on our relationship to AI safety:

We ...

15. heyshrutimishra (Group Score: 202.1 | Individual: 33.9)

Cluster: 7 tweets | Engagement: 189 (Avg: 100) | Type: Tech

You can turn a prompt into a game or 3D model with GPT-6 Astra on OnSolo.

I asked the OnSolo team for credits so a few of you could try it on your own ideas. We’ve got 100 credits each for 5 people here.

To enter, like the original OnSolo post below and leave a comment about what you’d build.

I’ll pick 5 people and collect your emails privately so the OnSolo team can add your credits.\n\nQT @OnSoloAI: GPT-6 Astra is now LIVE on OnSolo 🔥🔥 Multiple major updates just dropped, all powered by GPT-6 Astra. Pick your lane and start creating.

1️⃣ Game Section (Members only) Feed in a game idea. Agent spins up a playable web game — one click. 🎁 7-day limited free trial. Canvas usage limits apply: Premium · 1 use/day | Super · 2 uses/day | Ultra · 3 uses/day

2️⃣ Whitebox Video Expert (Free for all · Limited time) Canvas > Expert > Whitebox Video Expert Describe a scene in plain words. A box sliding down a ramp. A camera pushing through a doorway. Astra reasons out the space and the motion, then hands you a whitebox pass you can block your shots against. Previs before you spend a single render on the real thing.

3️⃣ 3D Model Expert (Free for all · Limited time) Canvas > Expert > 3D Model Expert Type what you imagine. A cute robot. A modern lamp. A stylized helmet. Astra builds the geometry, figures out the form, and delivers a 3D model ready to render, remix, or export. From idea to mesh in one prompt.

➡️What's next? One-click 3D scene modeling. One-click editing. One-click PV/MV generation. We're just getting started — and GPT-6 Astra is only warming up.

Ready to turn your ideas into reality? Open OnSolo and start building OnSolo.

@openai #gpt6astra #OnSolo #OnSoloAI

See 6 related tweets

@Parul_Gautam7: GPT-6 Astra is now LIVE on OnSolo !

The 3D Model Expert and Whitebox Video Expert are especially in...

@FellMentKE: GPT-6 Astra is live on OnSolo, and I’m interested in tools that shorten the distance between a rough...
@alex_prompter: What would you make with OnSolo? 👇

Reply to the original @OnSoloAI post below with your scene or ga...

@xIrissy: GPT-6 Astra is now LIVE on OnSolo 🚀

I’ve been trying it out and it’s exciting to see another powerf...

@heyrobinai: this might be the last major AI jump you get to catch early

GPT-6 Astra is so far ahead that every ...

16. rohanpaul_ai (Group Score: 198.5 | Individual: 62.5)

Cluster: 5 tweets | Engagement: 947 (Avg: 72) | Type: Tech

U.S. Vice President JD Vance had some unusually strong words for Anthropic on the All-In Podcast.

If an AI company develops technology that can be used for harmful purposes, such as cyberattacks, he believes that company has an obligation to help others defend themselves against those capabilities rather than relying primarily on government regulation.

"Why is it that the people who are at the frontier of the AI economy are throwing up their hands and saying, “Well, we’ve built Frankenstein,” and the solution to Frankenstein, apparently, is to create a one-world governance structure for artificial intelligence?

What I would say to those people is: if you’re building Frankenstein, stop. Or maybe, if the cat is out of the bag, then build the defensive mechanism against Frankenstein.

At the same time, that you have this sort of cyber-hacking tool that’s come out of Anthropic’s newest models, you have companies that are desperate for the defensive mechanisms to defend against that cyber-hacking tool, and they’re being denied access to it.

So, if you’re going to create Frankenstein, don’t come to the government and say, “We need regulation.” Look inward and accept that if you’re building Frankenstein, number one, you should stop.

And number two, when the companies come to you and say, “We need the tools to fight back against Frankenstein,” give them those tools."

full video link in comment on @theallinpod\n\nQT @theallinpod: All-In Summit: Vice President JD Vance 🔥

-- Assessing the Admin So Far

-- AI Approach, Dealing with Doomerism

-- Going After H-1B Abuse

-- Anti-Fraud Plan, Why Entitlements Fraud is Hard to Stop

-- Relationship with Israel, Rational Foreign Policy

-- Midterms Message

(0:00) VP JD Vance joins The Besties!

(1:40) Assessing the Trump Admin so far: Border control, re-industrialization, inflation, Iran, government debt, anti-fraud

(9:28) AI approach: Dealing with Doomerism, "Frankenstein," and AI negativity

(16:23) H-1B abuse and the massive $1T+ of fraud in federal spending

(21:48) Relationship with Israel, rational foreign policy

(25:19) JD Vance's midterms message

See 4 related tweets

@rohanpaul_ai: RT @rohanpaul_ai: U.S. Vice President JD Vance had some unusually strong words for Anthropic on the ...
@ShakeelHashim: “If you're building Frankenstein, stop.”

A very different tone to Trump here, while still calling f...

@AISafetyMemes: JD Vance: "Everybody I know tells me that Dario is very earnest, that this is a deeply held belief. ...
@Hesamation: > step 1: create problem

> step 2: demand solution https://t.co/ctxccFL62U\n\nQT @theallinpo...

17. daniel_brook_ai (Group Score: 190.4 | Individual: 28.2)

Cluster: 8 tweets | Engagement: 94 (Avg: 130) | Type: Tech

I’ve been using AI for more than just generating content.

These simple workflows can make everyday tasks much faster 👇

🔎 Research: Perplexity → ChatGPT → Claude

✍️ Content: ChatGPT → Claude → Canva

🎥 Video: ChatGPT → ElevenLabs → Runway → CapCut

📚 Learning: PDF → NotebookLM → ChatGPT

⚡ Automation: AI → Make/Zapier → Google Sheets

You don’t need 50 AI tools. Just connect the right ones and let them do more of the work. 🚀

Follow : @daniel_brook_ai

Which workflow would you try first? 👇

See 7 related tweets

@Zayan5754: RT @stevencoder9: 🔥 9 AI TOOLS EVERY CREATOR SHOULD KNOW

From design and presentations to research,...

@ElizabethA77617: RT @jonson_coder: 🔥 20 AI Tools You’ll Want to Know in 2026

AI is moving fast. These tools can help...

@lihazadn_Ai: RT @Shahriar661731: Last month, I was spending 5–6 hours a day doing work that AI can now do in minu...
@daniel_brook_ai: RT @CooperTechHub: 🚨 Stop using AI the hard way.

I found 120+ AI tools that can save you hours ever...

@robertsmith_ai: RT @heyfatema: 12 AI tools that can make your work feel 10x easier.

Claude → Think, write, brainsto...

18. jenzhuscott (Group Score: 183.8 | Individual: 33.8)

Cluster: 7 tweets | Engagement: 6 (Avg: 492) | Type: Tech

Set this one up for my hair lady. 😅 She sent me this & slightly mad at me for the overwhelming inflow. 😅 I love it when small business owners who use their hands and craft to earn an honest living get a boost from good AI services, especially those who are in my support system to keep my life functional. 💙\n\nQT @ElevenLabs: Introducing Reception, an AI receptionist platform for small businesses, built on ElevenAgents.

Every missed call could be a lost customer. Reception answers every call, answers questions, books the job, and texts confirmation.

Set up in minutes just by adding your website. https://t.co/95CNaBVH3X

See 6 related tweets

@Av1dlive: Wait… call centers and receptionists are SHAKING right now.

ElevenLabs just launched Reception[dot]...

@altryne: With @Muse and Instinct both launching "Agent calls businesses" mode, Eleven lab launches an AI rec...
@heyrobinai: a phone rang at 2am.. the AI picked up, answered the question, and booked the job

that's what Eleve...

@coreyganim: Never been easier to set up AI receptionists for local service businesses.

Easiest $5,000 you could...

@dr_cintas: This is one step ahead of Grok and Instinct.

AI workers with a full professional identity... we're ...

19. TFTC21 (Group Score: 183.4 | Individual: 38.6)

Cluster: 10 tweets | Engagement: 347 (Avg: 49) | Type: Tech

Elizabeth Warren's solution to everything she doesn't understand: ban it.\n\nQT @SenWarren: Today, I'm calling to immediately press pause on the development of advanced AI while lawmakers and regulators put systems in place to keep people safe.

See 9 related tweets

@jacob_wendler: The biggest takeaway at today's @politico Decoded Summit? Lawmakers and industry executives agree th...
@hollyotterbein: Some lawmakers argued that they don't need to use AI to write legislation about it.

“You don't need...

@danielnewmanUV: Great idea. Bring the economy to a screeching halt. 🤪\n\nQT @WatcherGuru: JUST IN: 🇺🇸 Senator Elizab...
@hollyotterbein: New: Members of Congress and governors are under increasing pressure to regulate AI, but more than t...
@WatcherGuru: JUST IN: 🇺🇸 Senator Elizabeth Warren calls to "pause" the development of advanced AI. https://t.co/g...

20. tetsuoai (Group Score: 177.5 | Individual: 51.2)

Cluster: 5 tweets | Engagement: 2609 (Avg: 554) | Type: Tech

SpaceXAI engineer Lingxi Li, @lingxi, runs code cleanup while everyone else is asleep. Nobody is shipping at 3 AM, so conflicts are rare and the changes are low risk.

His bots start a research agent that sweeps the monorepo for quality issues, poor modularization, comment bloat and security gaps. He wakes up to a stack of PRs.\n\nQT @bot: Three SpaceXAI employees are building a company in 3 days with Grok Bot. This is Day 1.

Matt Palmer (@mattyp), Lauren Tan (@poteto), and Roshan Sadanani (@roshan_s) start with research, a plan, and a product.

Live now, plus sessions for engineering, product, and founders.

https://t.co/eV3Tfd5Bf2

See 4 related tweets

@tetsuoai: SpaceXAI engineer Lauren (poteto) on why your codebase is a form of memory for agents.

The architec...

@XFreeze: Here’s how Grok Bot turns a simple idea into a fully working system

A SpaceXAI engineer builds a Gr...

@XFreeze: The biggest shift happening with Grok Bot: it doesn’t just help with tasks....it actually owns the o...
@elonmusk: RT @tetsuoai: SpaceXAI engineer Lingxi Li, @lingxi, runs code cleanup while everyone else is asleep....