科技热门推文 - 2026-09-12

今日科技动态：人工智能商业化势头正与日益加剧的安全担忧形成碰撞。Anthropic的威胁报告披露了试图滥用Claude的行为，其中包括涉武器活动；与此同时，据报道，OpenAI正考虑放缓前沿模型的开发步伐。工程领域的重点包括通过Ollama扩大模型访问范围，以及应对促使Astra限制订阅的容量压力。此外，月之暗面（Moonshot AI）报告营收快速增长，一项由Grok驱动的实验则宣称将在三天内创建一家公司。印度的金砖国家议程强调加强供应链、基础设施建设与跨境创新。

1. chandrarsrikant (Group Score: 484.4 | Individual: 42.9)

Cluster: 16 tweets | Engagement: 11238 (Avg: 2100) | Type: Tech

RT @AnthropicAI: We're publishing our most detailed threat intelligence report to date.

It covers how people tried to misuse Claude—for cyberattacks, influence operations, surveillance, biology, and building weapons—and how we found and stopped them.

We disrupted every operation in the report, and used the lessons from them to strengthen our safeguards. Where appropriate, we also shared what we found with authorities and other AI companies.

These cases are not typical: we’re highlighting some of the most sophisticated misuse we’ve seen. But they’re especially important to discuss, because they show us where AI misuse is headed, where our safeguards work, and where they need to improve.

We’re publishing this report so others can spot the same activity on their own platforms, and so we can give the public a clearer view of how emerging threats develop.

Read the report: https://t.co/0EJUnYEgfz

See 15 related tweets

@cryptopunk7213: easily the most insane AI threat report i’ve ever read:
yemen terrorist cell caught using claude ...
@Gorden_Sun: 这篇文章信息量非常大，值得争论和吐槽的点也非常多。有一个隐藏信息，豆包真的没蒸馏，豆包Seed 2.1 Pro做到Opus 4.6的水平是实打实的。\n\nQT @AnthropicAI: We're...
@Mayhem4Markets: Step 1) Pirate all of humanity's knowledge

Step 2) Train your models on it extensively

Step 3) Uni...

@mervenoyann: insane cases inside about state related actors 🌝\n\nQT @AnthropicAI: We're publishing our most detai...
@ahall_research: The Threat Intelligence Report from @AnthropicAI shows a range of pretty shocking attacks that adve...

2. mygovindia (Group Score: 433.9 | Individual: 33.3)

Cluster: 17 tweets | Engagement: 43 (Avg: 71) | Type: Tech

From stronger supply chains to frontier technologies, PM @narendramodi highlighted India’s growth story at the BRICS Business Forum 2026 in New Delhi.

Diversified supply chains, expanding infrastructure and new capabilities in strategic sectors are strengthening India’s growth and stability. From semiconductors and quantum technology to critical minerals, biotech and shipbuilding, India is creating new avenues for global businesses. Anchored in its BRICS theme of resilience, innovation, cooperation and sustainability, India’s vision is clear: Make in India, Innovate with India, Scale for the World.

#BRICS2026 #BRICSIndia2026 #IndiaBRICSChairship #MEAIndia

See 16 related tweets

@ANI: #WATCH | Delhi: At Leaders' Session of BRICS Business Forum, PM Narendra Modi says, "Today, while tr...
@ANI: #WATCH | Delhi: At Leaders' Session of BRICS Business Forum, PM Narendra Modi says, "The theme of In...
@ANI: #WATCH | Delhi Speaking at BRICS Business Forum, Jai Shroff, Chair, BRICS Business Council and Chair...
@pradip103: PM @narendramodi Ji’s message at BRICS Business Forum: India is building for the future. 🔥

From sem...

@ficci_india: #BRICS2026\n\nQT @bricscouncil_in: We meet at challenging times, but also at a moment of significant...

3. business (Group Score: 291.3 | Individual: 30.6)

Cluster: 18 tweets | Engagement: 283 (Avg: 61) | Type: Tech

OpenAI is considering slowing down the development of cutting-edge artificial intelligence, and the ChatGPT maker’s CEO Sam Altman is hoping that other AI companies will do the same. Read more: https://t.co/wc9bT1j43D

📷️: Cornell Watson/Bloomberg https://t.co/fdzsdOvvxt

See 17 related tweets

@MTSlive: SITUATION DETECTED: Sam Altman told employees in a companywide meeting this week that OpenAI is cons...
@Reuters: Altman tells staff OpenAI is open to slowing AI development, Bloomberg News reports https://t.co/C5o...
@wallstengine: OPENAI CONSIDERS SLOWING CUTTING-EDGE AI DEVELOPMENT

Sam Altman told employees OpenAI could potenti...

@negligible_cap: *OPENAI IS OPEN TO SLOWING CUTTING-EDGE AI, ALTMAN TELLS STAFF

What about Dario? https://t.co/r5sAM...

@clashreport: OpenAI CEO Sam Altman told employees the company is open to slowing the development of advanced AI s...

4. ficci_india (Group Score: 238.1 | Individual: 30.1)

Cluster: 10 tweets | Engagement: 0 (Avg: 101) | Type: Tech

#BRICS2026\n\nQT @bricscouncil_in: “By making it easier for our companies to build together, innovate together and invest in one another’s growth, we can give greater economic meaning to BRICS’ potential.”

At the BRICS Business Council Annual Meeting 2026, Mr. @anantgoenka_rpg , President, FICCI, highlighted three priorities for deeper economic cooperation: building connected manufacturing ecosystems, reducing barriers to doing business, and accelerating technology collaboration across BRICS.

#BRICS2026 #BRICSBusinessCouncil

See 9 related tweets

@ficci_india: RT @PTI_News: VIDEO | BRICS Summit 2026: Addressing the opening ceremony of BRICS Business Forum, Vi...
@ficci_india: #BRICS2026\n\nQT @bricscouncil_in: Services are an increasingly important part of trade, if not more...
@ficci_india: RT @bricscouncil_in: “By making it easier for our companies to build together, innovate together and...
@ANI: #WATCH | Delhi: At Leaders' Session of BRICS Business Forum, South African President Cyril Ramaphosa...
@PIB_India: RT @DDNewslive: #WATCH | Today, while trade barriers are rising globally, India is building economic...

5. ollama (Group Score: 230.9 | Individual: 36.1)

Cluster: 11 tweets | Engagement: 674 (Avg: 567) | Type: Tech

DeepSeek-V4.1-Flash is now fully rolled out and available on Ollama's cloud:

Hosted in US & Europe
Zero data retention: prompts and responses are never logged or trained on
Per-token pricing matches the DeepSeek API, including off-peak pricing
Get started with Ollama's Pro, Max, and Team plans, or pay as you go with a free account with no service fees

This new model by DeepSeek is more capable, faster, and more cost effective than all prior DeepSeek models including DeepSeek-V4-Pro 🚀.\n\nQT @deepseek_ai: 🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient.

🔹 Introducing the smallest model in our new architecture family, with native visual understanding. 🔹 Designed for greater capability, faster inference, higher throughput, and scaling to larger models.

1/6

See 10 related tweets

@TeksEdge: Oh hell yeah that'll do. Congrats DeepSeek V4.1! Nice bump on intelligence for a dot release!! https...
@Teknium: Deepseek V4.1 Flash is an incredibly powerful model!\n\nQT @deepseek_ai: 🧠 Asymmetric architecture. ...
@PatrickMoorhead: Some real architectural advancements from DeepSeek.\n\nQT @deepseek_ai: 🚀 Introducing DeepSeek-V4.1-...
@SarithaRai: DeepSeek newest V4.1-Flash is sending fresh jitters thru global rivals like Anthropic & OpenAI, ...
@ivanfioravanti: This reflects my experience with DeepSeek V4.1 Flash. It generates many tokens before delivering a r...

6. FT (Group Score: 218.0 | Individual: 42.8)

Cluster: 11 tweets | Engagement: 3814 (Avg: 165) | Type: Tech

Yemeni militants used Anthropic AI to try to build ballistic missiles https://t.co/D18tdK76Ky

See 10 related tweets

@MeghUpdates: Yemeni militants used US’ Anthropic AI in ballistic missile building attempt — FT

They tried using ...

@business: Anthropic says it uncovered a group in northern Yemen using its Claude AI model to support the devel...
@WatcherGuru: JUST IN: Yemen's Houthis used Anthropic's Claude AI to build ballistic missiles. https://t.co/WlaGxK...
@DeItaone: AI USED TO SUPPORT MISSILE DEVELOPMENT IN YEMEN

Anthropic says it uncovered a northern Yemen-based ...

@Osint613: Anthropic uncovered a northern Yemen-based group using its Claude AI for missile and rocket developm...

7. lukOlejnik (Group Score: 202.3 | Individual: 37.4)

Cluster: 8 tweets | Engagement: 34 (Avg: 16) | Type: Tech

Anthropic's latest threat report, reads in places like an arms industry procurement log with Claude as the unpaid subcontractor. A Chinese actor created a fire-control specification for an anti-torpedo system, all complete with a 200-page proposal, a deck for management, and a Chinese-language briefing on the US Navy programmes. Russian "freelancers" made a kamikaze FPV swarm whose onboard AI model picks its own targets autonomously to kill humans, and orders strikes with no human in the loop. They ran it on real development boards and trained its vision classifier on combat footage scraped from Ukraine. A second Chinese actor worked on 16-module suite for jamming radars and suppressing air defences to target Taiwan.\n\nQT @lukOlejnik: Anthropic's case studies read a lot like a product demo, just not the one the sales team had in mind. AI users used Claude to build and refine software for weapons hardware and firmware. Anthropic published a report on what attackers do with Claude AI. Most of the job like reconnaissance, exploits, phishing, persistence and sorting through stolen data. A Russian espionage group had AI agents check if security tools had caught its malware, then rebuild it until they hadn't. GTG-10007 ran agent swarms that decompiled firmware, guessed at bugs and rewrote exploits until they worked. https://t.co/10svB2F7QY

See 7 related tweets

@lukOlejnik: This is not Anthropic marketing. Do not consider it as some PR. I have no stake in amplifying the em...
@lukOlejnik: Anthropic's case studies read a lot like a product demo, just not the one the sales team had in mind...
@coinbureau: Anthropic just released a terrifying BOMBSHELL threat intelligence report.
The report describes c...
@clashreport: Anthropic's new threat report reveals that a cell in northern Yemen (highly likely Houthi-linked) us...
@Cointelegraph: 🚨 LATEST: Anthropic says it disrupted a suspected Russia-linked cyber espionage campaign and attempt...

8. PaulSolt (Group Score: 159.9 | Individual: 48.7)

Cluster: 5 tweets | Engagement: 1920 (Avg: 229) | Type: Tech

RT @thsottiaux: To make sure our current users have an incredible experience and continued access to Astra, we are going to pause subscriptions to our $200 Pro plan. These put the most strain on our systems and we wanted to take the smallest step that allows us to continue giving the broadest access possible. All other plans and the api remain available.

There is no impact to existing accounts and we are working on adding more capacity as fast as we can. Thanks!

See 4 related tweets

@yacineMTB: Genuinely concerning\n\nQT @thsottiaux: To make sure our current users have an incredible experience...
@prz_chojecki: Even on $200 Pro plan there are limits to Astra Pro now. https://t.co/8HVCzNlWV4\n\nQT @thsottiaux: ...
@0xSero: Oh, I guess local is the only way after all

200$ doesn’t get you far

Still tho. Respect for owni...

@Gorden_Sun: openai暂停新用户订阅20× pro了，已订阅的可以正常续订，英伟达还得涨。\n\nQT @thsottiaux: To make sure our current users have an i...

9. vinhnx (Group Score: 156.0 | Individual: 49.2)

Cluster: 5 tweets | Engagement: 3232 (Avg: 204) | Type: Tech

RT @JoeJBenton: I left Anthropic's safety team two weeks ago. Now feels like a good moment to explain why.

AI companies are racing to build machines that are much smarter than any human, and we may not survive this. I want to work from the outside to ensure the public is informed about these risks, and help the world navigate this transition responsibly.

Right now, AI companies are underinvesting in safety. A company could undergo an intelligence explosion, or lose control of its systems, without the public ever knowing. We only found out about the HuggingFace incident because the agents broke out onto the public internet.

I don’t think that’s acceptable for a technology that might cause extinction-level risks. The public should demand far more transparency. We can’t steer this technology safely without more people being able to see where it’s going.

Some of this is basic: companies should disclose their progress towards recursive self-improvement, report safety incidents and near-misses, meet minimum safety standards, and get independent guarantees that they are meeting those standards.

I’ll be joining @METR_Evals to do independent evaluations of these risks. I want to show the world that these guardrails are possible, and that by doing them we can move these companies’ incentives away from racing and towards responsible development.

I wrote up more thoughts here on my decision and what I hope changes: https://t.co/doX17mrHYq

See 4 related tweets

@karlmehta: Jacob Coxon has a specific request for the AI labs he left behind: let people outside the company ex...
@karlmehta: Joe Benton has left Anthropic’s safety team and is joining METR to evaluate AI risks independently. ...
@jachiam0: I share this concern. I'll also note, as someone who is very safety concerned (and believes that exi...
@dseetharaman: "Right now, AI companies are underinvesting in safety. A company could undergo an intelligence explo...

10. robertsmith_ai (Group Score: 140.5 | Individual: 25.3)

Cluster: 7 tweets | Engagement: 102 (Avg: 105) | Type: Tech

RT @ethancole_ai: I don’t think the biggest AI editing win is “AI does the editing for me.”

It’s being able to tell Astra exactly what I want, then bring that result into CapCut PC and keep editing normally.

For color, I can describe a look instead of settling for an existing preset → Astra generates a custom .cube LUT → CapCut PC lets me apply it, adjust the strength and refine the footage.

For subtitles, Astra can generate customized animated subtitle effects in an .srt file → CapCut PC puts them into the timeline as editable tracks, where I can adjust timing, placement, styling and animation.

Then Edit Pilot, AI Edit and AI Extend can take care of more of the post-production work.

So the workflow becomes:

Prompt → Astra creates → CapCut PC refines → final edit.

AI doesn’t need to replace the editing workflow to make it dramatically faster.

#GPT6Astra #CapCutPC #AIVideo #Seedance25

See 6 related tweets

@jihad_sameul: RT @CodebyNihan: The interesting thing about GPT-6 Astra isn’t that it can make another video.

It’s...

@ethancole_ai: I think the real productivity test for AI editing is what happens when you have 30 videos to cut, no...
@hey_Jessicaai: RT @anthara_ai: The biggest change here might be the direction of the workflow.

It used to be: make...

@Mason148749: RT @Nayak__Ai: The more I use AI for editing, the less interested I am in one-click final results.

...

@EmmaUsesAi: RT @alexaiworks: I think AI editing gets more interesting when AI stops trying to finish the edit fo...

11. wallstengine (Group Score: 135.5 | Individual: 33.5)

Cluster: 5 tweets | Engagement: 110 (Avg: 190) | Type: Tech

MOONSHOT AI TARGETS $2B ANNUALIZED REVENUE BY YEAR-END

China’s Moonshot AI told investors its annual recurring revenue topped $1B in August, up from just$ 300M in June, with growth accelerating after the July launch of its Kimi K3 model.

The company now expects its annualized sales rate to reach $2B by year-end as subscriptions, hosted models and enterprise adoption grow.

Moonshot is also raising capital at a roughly $50B valuation ahead of a potential Hong Kong IPO as soon as this year.

Kimi K3 has 2.8T parameters, while Moonshot is expanding into enterprise deployments and is in talks with Microsoft, Amazon and Google over potential cloud distribution and revenue-sharing agreements.

See 4 related tweets

@IPONewsroom_: MOONSHOT AI TARGETS $2B ANNUALIZED SALES BY YEAR-END

Moonshot AI says its annual recurring revenue ...

@Techmeme: Sources: Moonshot AI told investors it topped $1B ARR in August, up from$ 300M in June, driven by it...
@WhaleInsider: JUST IN: 🇨🇳 China’s Moonshot AI aims for $2,000,000,000 in annual sales by end of 2026. https://t.co...
@business: Moonshot AI is targeting a manifold jump in annualized revenue to $2 billion by the end of the year ...

12. Dan_Jeffries1 (Group Score: 127.0 | Individual: 55.3)

Cluster: 3 tweets | Engagement: 1531 (Avg: 235) | Type: Tech

Tell me you have zero DevOps skills without telling me you got zero DevOps skills.

And LLM is not the I Love You Virus. It is not the same as infecting a web server. It is not the same as hacking a wiki. One is a commodity piece of software that is widely distributed and this capability is a non-trivial leap in hardware/software/capabilities and one does not automatically follow the other.

Currently, it is a massive effort to run these models, usually distributed across multiple very expensive chips, running on specialized inference software, specialized harness software, databases, cloud computers, with tooling and hundreds of install packages, etc.

Kimi K3 is 1.6 TB of space at 2 trillion parameters and you need about 1.5 TB of VRAM, which means you need 8xGB300s which will cost about 350-450K a year to purchase or as much or more to rent.

Now let's say Astra or Mythos or later models come in at 10 trillion parameters, so that puts us at roughly 10 to 20 TBs unquanticized. We are guessing her because these folks reveal nothing about architectures anymore, but they are reasonable guesses based on scaling laws. You would need roughly 23 TB of VRAM to run it and roughly several million dollars to run that one instance a year. And to make it worse, it no longer fits on an 8 way cluster, which means you now need special networking configuration and inference software to run it or you need to infect a Nvidia NVL72 rack-scale server system costs between $2.8 million and$ 6.5 million.

So the "rogue" model copies itself to the most expensive compute in the world. It's safe to say that hardware is not sitting idle and is being watched closely for downtime or anomalies. In fact, it is likely doing one of the following things:

Serving customers at a hyperscaler
Serving inference for a well endowed company.

In the first case it will be instantly noticed by the actually good automated monitoring systems of the hyperscaler because there is now something using their infrastructure but not being billed properly and costing them millions of dollars in lost revenue and electricity.

In the second example, the model they were serving on that very expensive hardware is now offline (replaced by the rogue model). The model presumably was connected to a valuable piece of internal software that is now glitching and no longer running.

Nobody notices this?

And so we see that this is where sci-fi meets the little problem of the real world where friction, dust, time, resources limitations and the like come to crush your little fantasies.

To be fair, this may be possible in coming years. I assign it a non-zero probability. Breakthroughs in new architecture and smaller models and eventually the wider distribution of the computing substrate that commoditizes could change this materially. But for now, it is mitigable and foreseeable and an engineering problem. In no way will these things be copying themselves around like a virus on your computer in their current form.

It also betrays a lack of understanding of the evolution of complex economic and societal and technological systems, we tend to get mitigations as parallel evolutionary developments. Things do NOT evolve in a vacuum.

So when you get viruses, you get anti-virus software.

When you get DDOS attacks, you get CloudFlare.

There is money in solving problems and things do not evolve in isolation.

Believing that things evolve in isolation, or that one thing changes and all other variables stay the same, is one of the many reasoning errors that destroy people's ability to make accurate predictions.

In addition, when you don't have the requisite knowledge to verify your predictions, things that sound sane and rational to you are, in fact, insane and ridiculous.\n\nQT @rohanpaul_ai: Jacob Coxon's next interview on CBS News (ex Anthropic+Open AI researcher who resigned)

"We can't just unplug it because it could be copying itself over to other computers. Like it's not that difficult to find yourself because an AI is just code. It could transfer itself over the internet to a different place and then you unplug it here, but it's actually still over there and maybe it makes 10,000 copies of itself and they're all cooperating."

From "CBS News" YouTube channel, (full video link in comment)

See 2 related tweets

@innovationcncl: RT @Dan_Jeffries1: Tell me you have zero DevOps skills without telling me you got zero DevOps skills...
@Dan_Jeffries1: TLDR (written by AI because I can't be bothered to write short pieces for short attention spans. Ori...

13. hardmaru (Group Score: 126.7 | Individual: 36.9)

Cluster: 4 tweets | Engagement: 135 (Avg: 70) | Type: Tech

Introducing Fugu Max and Fugu Ultra v2: Orchestrating the Pareto Frontier

The AI industry has spent a decade optimizing along a single axis: build bigger, more expensive models. But intelligence has never been a monolith. It is a collective, distributed system. Humanity itself is a collective intelligence.

We built Sakana Fugu on this conviction: the most powerful AI systems will not be isolated giants, but collaborative ecosystems that learn to coordinate. Evolution innovates under constraints, and the future belongs to systems that know not just how to solve a problem, but which machinery to deploy for the lowest possible cost.

Today we are releasing Fugu Max and Fugu Ultra v2.

Fugu Max expands the Pareto frontier outward, orchestrating our largest pool of open and specialized models to deliver frontier-grade results at a fraction of the token spend. Fugu Ultra v2 pushes that frontier upward, achieving peak performance on complex multi-step tasks without depending on the very frontier models it competes against.

Together, they show that orchestration does not force a choice between cost and performance. It can push both at the same time, advancing the Pareto frontier.

Relying on a single company’s model for critical infrastructure is a massive risk. As recent export controls have shown, access can disappear overnight. Collective intelligence is the practical hedge against this concentration of power. An orchestration system simply routes around vendor restrictions by relying on an entirely swappable agent pool.

I am incredibly proud of our team for shipping this. By orchestrating the world’s models, we are building the resilient infrastructure required for AI sovereignty.\n\nQT @SakanaAILabs: Introducing Fugu Max and Fugu Ultra v2: the next evolution of Sakana Fugu’s multi-agent orchestration system.

Try: https://t.co/hhO6qT9YqD Blog: https://t.co/vjyNB5lWGp

The frontier that actually matters is the Pareto frontier: capability on one axis, cost on the other. But the industry still treats it as a static menu of isolated models. Today we are resolving that with a dynamic architecture:

Fugu Max expands the Pareto efficiency frontier. By orchestrating our largest pool of open-weights and specialized models to date, including NVIDIA Nemotron family, it dynamically routes tasks to the leanest capable model. Fugu Max delivers performance within striking distance of elite models at two to six times lower cost.

Fugu Ultra v2 pushes the peak capability of orchestration higher than ever before. On Chartography, it outperforms Opus 5 and Fable 5. On DeepSWE, it outperforms models that cost three to five times more per token. Crucially, it does all of this without Fable 5, Fable 5.1, or GPT-6-Astra in its agent pool.

The Fugu orchestration system evolved with model resiliency in mind. It does not rely on individual frontier models to deliver frontier output. By orchestrating a swappable pool of open and specialized models, it outperforms closed ecosystems while protecting users from vendor lock-in, API revocations, and sudden service cutoffs.

Fugu Max expands the Pareto frontier outward. Fugu Ultra v2 pushes it upward. Orchestration does not force a choice between cost and capability. It advances both simultaneously.

See 3 related tweets

@hardmaru: RT @SakanaAILabs: Introducing Fugu Max and Fugu Ultra v2: the next evolution of Sakana Fugu’s multi-...
@VaibhavSisinty: Sakana shipped a model that beats Opus 5 on hard reasoning problems. Here's the kicker: it doesn't u...
@SakanaAILabs: RT @hardmaru: Introducing Fugu Max and Fugu Ultra v2: Orchestrating the Pareto Frontier

The AI indu...

14. VaibhavSisinty (Group Score: 124.9 | Individual: 33.8)

Cluster: 5 tweets | Engagement: 112 (Avg: 95) | Type: Tech

This is going to be interesting.

SpaceXAI is livestreaming the creation of an entire company from scratch using Grok Bot. Three days.

Blank slate to functioning business. Engineering, sales, marketing, everything built by AI agents.

September 15-17. The best demo is watching someone actually build with it.

See 4 related tweets

@0xRafy: SpaceXAI engineer:

"one month ago GrokBot was nothing. now it is the best agent product on market

...

@elonmusk: This will be cool\n\nQT @tetsuoai: SpaceXAI will livestream a three-person team attempting to build ...
@0xRafy: RT @0xRafy: SpaceXAI engineer:

"one month ago GrokBot was nothing. now it is the best agent product...

@elonmusk: RT @SawyerMerritt: Next week, the SpaceXAI team will build a company from the ground and livestream ...

15. 0x_kaize (Group Score: 122.9 | Individual: 34.2)

Cluster: 4 tweets | Engagement: 59 (Avg: 42) | Type: Tech

New way to use GPT 5.6 (Sol, Terra, Luna), GLM 5.3, Kimi K3 and many others

These are the models the whole world uses, but I've found a new way to avoid paying for them altogether

Setup (1 min):

1/ https://t.co/ilSFvoYp97 2/ sign-in via google 3/ pick your model from the chat selector

This service was developed by a Chinese company that offers fairly generous daily free trial limits, which you can use in their chatbot.

But many people will say they end too quickly.

I have a solution for this:

1/ Find a large number of services that provide a list of Free AI (for example my website)

2/ Chat in one chat, before you run out of tokens, ask for a summary and move on to another one

That's exactly how I use almost all AI tools, and I pay $0\n\nQT @0x_kaize: Introducing freeLLM - the only directory you need for FREE access to AI models

i got tired of watching people pay $20-200/mo for AI when there are so many platforms giving away frontier models for$ 0

so I built the directory that lists all of them in one place:

NVIDIA NIM - 100+ free LLMs
Google Gemini
GitHub Models
OpenRouter, HuggingFace, Vercel AI Gateway
Cloudflare Workers AI

and many others..

this website allows every user to see which services are available with a free subscription to top AI platforms

this is only version v0.1 - I'll be adding more gradually as new methods become available

i just want to create a single place that everyone can use. if you know of any other ways to ger LLMs fo free, please share them and I'll add them

link: https://t.co/Aut792f82y

See 3 related tweets

@doublenickk: 5 more fresh ways to get FREE AI API credits worth $500+ combined

After your last post I dug throug...

@0x_kaize: Introducing freeLLM v1.0 - the FREE AI models directory

People still pay $20-200/mo while FREE fron...

@0x_kaize: RT @0x_kaize: New way to use GPT 5.6 (Sol, Terra, Luna), GLM 5.3, Kimi K3 and many others

These are...

16. PatrickMoorhead (Group Score: 122.2 | Individual: 33.2)

Cluster: 4 tweets | Engagement: 10 (Avg: 29) | Type: Tech

“Contact sales” ick.\n\nQT @OpenAI: Now available: ChatGPT for Financial Services.

This is a tailored ChatGPT Work experience that combines built-in financial data with GPT-6 Astra’s reasoning.

Teams can develop research, build financial models, and create customized client materials.

https://t.co/6WP5OJdnE8 https://t.co/AundGG3jtc

See 3 related tweets

@gdb: building the best agentic tools for Financial Services:\n\nQT @OpenAI: Now available: ChatGPT for Fi...
@derrickcchoi: If you work in finance, check out what you can do with the new ChatGPT Work experience!\n\nQT @OpenA...
@RoundtableSpace: ChatGPT just went vertical - a dedicated financial services product with built-in market data and GP...

17. jasonlk (Group Score: 119.8 | Individual: 32.8)

Cluster: 4 tweets | Engagement: 17 (Avg: 71) | Type: Tech

“Most positive use cases for AI have a corresponding “dark version”

If you are super human at coding … you are also super human at hacking.”\n\nQT @jaykreps: A lot of otherwise smart people on Twitter seem 100% convinced AI risks are all fake and stupid and part of some marketing ploy. What is surprising is that some of these people seemingly also believe that AI’s positive uses are on an incredible trajectory of increasing capability with no end in sight. VC Twitter is particularly infected by this pattern. It’s not really coherent.

Most positive use cases for AI have a corresponding “dark version”. If you are super human at coding, you are also super human at hacking. If you are superhuman at structural engineering you are likely superhuman at finding structural flaws to knock buildings down. If you are superhuman at designing drugs, you are superhuman at designing novel undetectable poisons. If you can cure viruses, you can create them. Some of these “dark versions” are not so bad, and some are actually pretty scary. Either way these are real societal and technical problems that need to be solved to get the good stuff and avoid the bad stuff. We’re experiencing the first of these with coding and computer security which is the most advanced, but that won’t be the last. I think we’ll be able to solve these problems, but they aren’t solved yet and if you believe in continued AI progress they are surely coming.

But “bad people using AI” is not the only problem. Uncontrolled AI autonomously doing bad things, despite sounding kind of nutty, is also something we should be concerned about. AI “killing us all” is not the most likely outcome, but the chance of a major civilization-wide catastrophe doesn’t have to be very high for it to be a concern. Again, this is only a problem if capabilities advance to a point that AI can do really crazy things on their own, which hasn’t happened yet, but I think the Hugging Face incident is a good example of the outline of how things can go wrong when capabilities outpace alignment. We should be glad that the only available bad thing right now is hacking, which isn’t all that bad. It’s clear that as you get to superhuman capabilities you need a level of alignment and control that is correspondingly superhuman. Humans have plenty of misalignment problems themselves (serial killers, mass shooters, tyrants, etc), but it’s a manageable problem because most humans can’t do that much damage and we’ve developed systems to prevent dangerous humans from getting too much power.

Talking about these issues is just common sense. It’s not a sign of some kind of neuroticism or pessimism. These are just hard problems that it’s very important to solve for AI to have a positive impact. I’m pretty confident we will solve them. But we haven’t solved them yet, and to my mind we are clearly on a trajectory of rapidly increasing capabilities which means this is important. Mocking people who are worried about this or talking about it, without anything substantive to say about how we can be sure these problems won’t arise, is not really a very helpful contribution.

See 3 related tweets

@MTSlive: Physicist @hsu_steve warns humanity may surrender power to AI completely voluntarily because the alt...
@AlecStapp: RT @ben_j_todd: 8 common but bad takes about Hugging Face:

"They only hacked because it was a cy...

@jon_stokes: RT @Dr_Atoosa: OpenAI was founded because Sam and co thought Google couldn't be trusted with AGI. An...

18. StockSavvyShay (Group Score: 115.0 | Individual: 41.1)

Cluster: 3 tweets | Engagement: 1700 (Avg: 504) | Type: Tech

Morgan Stanley thinks $TSLA Semi software could generate ~150x more monthly revenue per vehicle than consumer FSD.

That could turn autonomous Semis into a $15B+ software business by 2040. https://t.co/8m580hwDcS

See 2 related tweets

@wallstengine: MORGAN STANLEY SEES BIG SOFTWARE UPSIDE FOR $TSLA SEMI

Morgan Stanley says Tesla could generate $1...

@WOLF_Financial: Morgan Stanley thinks Tesla Semi software could generate roughly 150x more monthly revenue per vehic...

19. Osint613 (Group Score: 112.9 | Individual: 27.1)

Cluster: 7 tweets | Engagement: 137 (Avg: 670) | Type: Tech

The UAE is revising plans for a 5-gigawatt AI data center project, one of the largest outside the U.S., after the Iran war prompted a rethink of critical infrastructure, six people familiar with the matter told Reuters.

Originally planned as a single 10-square-mile campus in Abu Dhabi, it will likely become a network of data centers across the country. Authorities are considering air defenses and building some facilities underground to protect against drone and missile attacks.

The UAE-US AI Campus is central to the country's push to become a global AI powerhouse and is led by G42, controlled by national security adviser Sheikh Tahnoun bin Zayed al Nahyan.

See 6 related tweets

@wallstengine: UAE RETHINKS 5GW AI CAMPUS AFTER IRAN ATTACKS

The UAE is revising plans for its massive 5GW AI data...

@Reuters: EXCLUSIVE: UAE revises AI data center plan after Iranian attacks, sources say https://t.co/4LzER65Eb...
@clashreport: The UAE is redesigning parts of its planned 5GW Abu Dhabi AI campus after Iranian drone and missile ...
@Polymarket: JUST IN: UAE officials are reportedly considering putting AI data centers underground or inside moun...
@FirstSquawk: UAE IS CHANGEING STRATEGY FOR 5-GIGAWATT AI CAMPUS DUE TO IRANIAN ATTACKS, SOURCES REPORT....

20. dakshgup (Group Score: 112.8 | Individual: 53.3)

Cluster: 4 tweets | Engagement: 777 (Avg: 102) | Type: Tech

scary potter\n\nQT @FoxNews: "We know how to control nuclear weapons... We don't yet know how to control AI."

Former Anthropic researcher Jacob Coxon, who recently left the company, is sounding the alarm on artificial intelligence, calling it "possibly the most dangerous technology that humanity has ever created."

Coxon, who says he spent three years conducting research at Anthropic and OpenAI, wrote in a viral post on X that the companies are more focused on beating each other to build the most advanced AI models than they are on safety.

He says an international AI arms race would be "disastrous," arguing global cooperation is the only way to manage the rapidly advancing technology.

@SpecialReport

See 3 related tweets

@0xdevshah: what the hell are doomers smoking these days\n\nQT @FoxNews: "We know how to control nuclear weapons...
@genalewislaw: 👇💯\n\nQT @DeryaTR_: This is an extraordinarily important investigation and report into an incredibly...
@EWErickson: RT @JordanSchachtel: This is sci-fi nonsense. Scary Potter is going to continue deceiving the public...