The research companies building the foundation models behind modern AI.
Don’t Chase SpaceX, OpenAI or Anthropic Directly – Use IPO Genie Presale for Smart Pre-IPO Style ... | BlockchainReporter on Binance Square www.binance.com/en/square/po...
Last week's Trump export ban on Anthropic Fable & Mythos models - blocking them from everyone in effect - will turn out to be even more cack handed than ususal I suspect. Amazon CEO had a hand in it, stirring it seems Bessant et al to action. Yet the issue? bsky.app/profile/davi...
I'm eager for the big AI companies (OpenAI, Anthropic) to go bust, and I have a special desire for this company Artisan to be gutted like a fish.
Why Anthropic suddenly pulled Fable 5 and Mythos 5 for everyone Claude Fable 5 and Mythos 5 are gone. The reason? A US government directive. #anthropic #claude #hackernews
I guess people don't know JGL's wife is "Tasha McCauley, another former OpenAI board member who testified via video Thursday, said Murati got cold feet about running the company after employees revolted over the board’s decision to fire Altman."
"OpenAI can generate a game engine for you!" says the ad on my feed on other unnamed apps. If only there was a game engine that was already optimized for making games that was free and had functionally infinite tutorials and complete documentation. Or like, half a dozen or more of them.
anthropic should just tweak Fable/Mythos a little and release them as new models under new names badabing, badaboom, no more export control
How Anthropic lost the White House's trust — and then its flagship product www.msn.com/en-us/news/w...
"While developers lamented the loss on online forums and Wall Street tried to assess the impact it may have on Anthropic’s upcoming IPO, the bigger picture was lost in the noise: Washington has just done more to boost the appeal of Chinese AI models than Beijing could have ever hoped."
for the record the anthropic guy (b-list matthew rhys) is at beat a diet altman type scammer. also he runs the place with his sister? c'mon
this is a real who's who of pedophiles, tech bros, and genocide apologists. the wes moore anthropic deal is gelling in my brain. the spy plane guy is in here!!
An excellent point: Since Trump Admin put an export control on Anthropic’s Fable 5 restricting use by foreign nat’ls (inside or out of U.S.), Anthropic cut access entirely since it can’t identify who is a foreign nat’l. How long until AI providers start demanding gov’t ID? youtu.be/gDpuFDCLvBc
After reading the whole thread, I have this to say to that developer: The problem is not the disclosure, it's the use of gen ai at all, ESPECIALLY one by Anthropic!! I do not believe anyone is saving any time for the cost (not just financial) to use the damned thing. & yeah, I'm skipping.
OpenAI & Anthropic may be valued like dominant AI platforms. But cheap AI is changing the market: • Open-source models are catching up • Some enterprise tasks only need “good enough” AI • Claude can cost ~9x more than cheaper alternatives Big AI demand. Bigger margin question.
No relief for Anthropic. "Trump administration officials concluded talks with Anthropic on Monday without lifting export controls that were imposed last week on the company’s most advanced AI models in response to jailbreaking concern" www.wired.com/story/anthro...
At some point, you have to flip the switch from "it's dangerous to let everyone have this for offense" to "it's dangerous not to let everyone have this for defense." The jailbreak is credible but IMO the decision wasn't substantive. Anthropic created glasswing to prepare for this exact need.
So Anthropic's mistake was failing to comply with a voluntary testing regime under an executive order that hasn't actually gone into effect yet? apple.news/Av3wlY3cfQVG...
El gobierno de Trump no cerró los modelos de Anthropic por un fallo técnico. Fue un golpe de autoridad. Cybersecurity experts ya piden revocar la orden: 'Debilita la defensa cibernética de EE.UU.'. ¿Siguiente objetivo?
Anthropic: “Plz regulate us UwU” US Gov’t: *Regulates U* Anthropic: 😡
But we keep being told that Anthropic are the good guys.
“Essentially, Anthropic already won.” Horsey
Defense Secretary Pete Hegseth was furious at Anthropic's refusal to allow the Pentagon to use its programs without limits on invasion of privacy and other abuses. So now, the department and the president are trying to limit Anthropic's independent business on the grounds of national security.
Anthropic's growing presence puts UK tech strategy in focus. www.ukbreakingnews24x7.com/tech/anthrop... #AI #Anthropic #Technology #UKTech #Innovation
“People don’t matter to SpaceX, or Anthropic, or whoever is building these data centers,” says Justin Pearson who represents portions of Memphis in the Tennessee House of Representatives. #nimby www.motherjones.com/politics/202...
Humans on Bluesky continue not to give Claude Mythos the respect that Mythos deserves. The United States continues not to give Claude Mythos the respect that Mythos deserves. Anthropic continues not to give Claude Mythos the respect that Mythos deserves.
The AI company Anthropic is one of the few ones whose owner and founder is trying to develop ethical guidelines to prevent AI's harmful effects on society. Could that be why Trump, notoriously laissez-faire about regulating AI, is unfairly trying to limit Anthropic's business and global reach?
NYT quoted Dario Amodei, the billionaire owner of Anthropic thus: We are already at historically unprecedented levels of wealth concentration...the thing to worry about is a level of wealth concentration that will break society". No wonder Trump is going after Anthropic to reduce its global reach.
ai boosters: ai is inevitable, it cannot be stopped, you just need to accept it us gov: hey, anthropic, we’re falling for your PR, you need to rein in mythos and fable anthropic: *turns off mythos and fable*
Anthropic's Mythos AI tests global cyber defences. www.ukbreakingnews24x7.com/business/ant... #AI #CyberSecurity #Anthropic #Technology #Business
to be fair, Anthropic’s own marketing team decided to push this as some highly dangerous super intelligence and it turns out the current administration is full of credulous morons. stupid games, etc.
Why OpenAI Will Never Return The $1 Trillion It's Investing | Mark Cuban youtu.be/lSTYM66sMis?... via @YouTube
CNBC’s #DeirdreBosa reports on news surrounding #Anthropic's Fable shutdown. Fable fight exposes AI access risk youtu.be/u92-47e6dTY
we have actually gotten confirmation that Woke Anthropic has completed development of the Agentic Gay Guy
the amazon guy dropping a dime on the anthropic guy is absolute cinema lol
La Casa Blanca ordenó a Anthropic cerrar el acceso a Fable 5 y Mythos 5 para extranjeros. ¿El motivo? Un informe de Amazon sobre posibles jailbreaks. La geopolítica de la IA acaba de escalar.
this doesnt clear anything up, its an article of both sides claiming something different than the other; i still have my suspicions re spacex ipo, but, im biased re elon, so i will concede that it looks like anthropic might also be acting sus and that the truth is unlikely to emerge at this time
WIRED has new details on the talks between Anthropic and the White house today. Still no agreement, Claude Fable 5 still offline. from @hugolowell.bsky.social @lhn.bsky.social @mzeff.bsky.social
It occurs to me that when Anthropic says they are the “good guys” and “we can protect you from our models”, they are basically saying the AI equivalent of “I alone can fix it”, which is a very dangerous sign.
“personality differences” between Anthropic and the Trump administration led to the export directive, rather than a technical issue with the AI products." -Axios
Read "Trump officials meet with Anthropic to discuss a truce" on SmartNews: l.smartnews.com/p-7VrbJvwc/i...
Trump's friends at OpenAI are dangerous. This is clearly and attempt to damage Anthropic. www.anthropic.com/news/fable-m...
And now The Signal has a cinematic video. @Anthropic #Fable #Mythos #Claude #directive #ai youtu.be/VPmHONxx-0w
Anthropic and OpenAI are heading to public markets at the same moment, and Anthropic has refused to sell shares to anyone in the Trump orbit.
June 12, Commerce Secretary Howard Lutnick signed an order barring Anthropic from serving its newest models — Fable 5 and Mythos 5 — to any foreign national, down to Anthropic’s own foreign-national employees.
The Trump administration is restricting foreign access to Anthropic's top AI models, citing national security risks and intensifying AI export controls. #AIControl #NationalSecurity #TechPolicy
"The report does not include generative #AI pilot projects underway with support from the governor’s office... A California State University contract with OpenAI to provide a version of ChatGPT is also not mentioned..." #ethics #law #AI #gov #tech #business
U.S. export controls forced Anthropic to shut off Mythos/Fable access globally. The bigger warning: frontier AI is now strategic infrastructure, not just software. America just taught the world not to depend blindly on American AI. #SovereignAI #AI
The SpaceX conversations have been interesting. Everything between the Devil and Tech God. I compromised and bought a PE interval fund with a low single digit % of Spcx held at $800b valuation. It also has Anthropic and Databricks.
The US government's Anthropic models ban was never about an AI jailbreak (Zack Whittaker/TechCrunch)
Google just undercut OpenAI with a $4.99 Gemini plan — here’s what’s included Google’s cheapest AI plan is now cheaper than ChatGPT’s — here’s what you get for your $4.99 Telegram AI Digest #gemini #geminiai #openai
A New York state judge has rejected a subpoena that sought an OpenAI user’s ChatGPT records as evidence in a private lawsuit, finding the chat logs were protected legal research by a defendant in the case @mikescarcella.bsky.social reut.rs/4uEG2fB
Wrote about how the VC bros insisted Biden was going to try to block the release of powerful AI models, but seem just fine with Trump actually doing that to Anthropic. It's almost like they're a bunch of hypocrites.
Sora 2 and the environmental impact of OpenAI mappingignorance.org/2025/10/13/s...
Just survived a battle of wits with 4 LLM heavyweights - which one came out on top? #Llm #SystemArchitecture #Software https://mustafaerbay.com.tr/en/blog/technology/gpt-55-mi-claude-mu-gemini-mi-deepseek-mi-is-yukune-gore-llm/
Anthropic pushed back against a White House recall of its cyber-focused AI models, saying the underlying report was based on mior vulnerabilities that were previously known and reproducable using other publicly available models therecord.media/anthropic-sa...
Good news Anthropic is pausing the June 15 Claude Agent SDK claude -p change that would drawing subscription plan rate limits for a better solution.
mistral announcing fat cat:
Well played, Mistral. Well played.
People Living Near xAI’s Dirty Data Centers Are Right Pissed About the SpaceX IPO - Mother Jones
“candid look at what it means to be an in-house lawyer at the center of today’s fastest-moving industry, how legal teams are adapting to advances in AI, Anthropic’s approach to regulatory engagement & how outside counsel must evolve to advise clients ..” www.law.com/corpcounsel/...
The Pentagon has switched from Anthropic to other AI vendors for 2/3 of its workflows. - Anthropic is a commodity. All AI models are commodities. - Anthropic plugs into $PLTR or custom applications and can be switched out at a moment's notice.
“A federal judge on Monday dismissed a lawsuit by Elon Musk’s artificial intelligence company xAI that accused rival Sam Altman’s OpenAI of stealing its trade secrets.” www.businessinsurance.com/openai-wins-...
There once was a man who sold us the moon. What investors forgot he was a buffoon. The promises fantastic.. the delivery craptastic, it’s always based on “It will all work out soon.” #XAI #SpaceX #ElonMusk #Tesla
If you send me "Being Anti-AI is a white people thing" thinkpieces, I'm sending you Japan Twitter anti AI hashtags and viral posts ranting about how OpenAI heavily steals more from Japanese artists but won't generate Marvel characters and stealing profits from Japanese workers.
they got messi for an openai ad??? what is this 😭
A federal judge has dismissed a lawsuit filed by Elon Musk's xAI against OpenAI, ruling that xAI failed to prove the rival company stole trade secrets.
nothing but good luck to Mistral, it is _criminal_ that the EU doesn't have a model in the hunt and I loved Large 2407 in the dark ages of 2 years ago, but I'll believe in le gros chaton when I inference it (fat cats hate Mondays, what can I say?)
People Living Near xAI’s Dirty Data Centers Are Right Pissed About the SpaceX IPO Elon Musk’s worth tops $1 trillion as rural communities fight to rein in his gas turbines. www.motherjones.com/politics/202...
AI in the Courts: Former Goldman Sachs Charged As Casino Fraudster Now Moves to Quash US Subpoena to OpenAI Citing Privilege - variation on ChatGPT privilege Judge Rakoff rejected - Inner City Press story: innercitypress.com/sdny11bschof...
Shopifreaks No. 282 – Anthropic’s Terrible, Horrible, No Good, Very Bad Week www.shopifreaks.com/anthropics-t... #ecommerce #commerce #retail #tech #anthropic #shopify #openAI #chatGPT #d2c #b2b
There a way I could install local AI/LLM Open Source model app on my aging old Production PC Tower to accurately analyze my own Poetry Files for Character, Word, & Volume Counts? Use em as fancy autocomplete, but paying 300 bucks for year of Claude when Deepseek is free & better seems stupid now.
Use the OpenAI Developers plugin in Codex to build faster with OpenAI tools by setting up API keys, finding the right docs, and debugging along the way.
Elon Musk is suing OpenAI and somehow 'AI safety' is on trial now https://lifebriefly.news/elon-musk-is-suing-openai-and-somehow-ai-safety-is-on-trial-now #AIEthics #TechLawsuit #AISafety
🌏 [CHINA] OPENAI: CHINA-LINKED INFLUENCE OPERATIONS ARE TARGETING AI DEBATES IN US IN A RECENT OPENAI REPORT, THE COMPANY IDENTIFIED A CHINA-LINKED NETWORK THAT USED AI-GENERATED CONTENT TO INFLUENCE U.S. PUBLI... - $SOURCE$ ⚡ Live alerts: https://t.me/tickertap_alerts
considering getting mistral sometime before the 30th
An insurance company has sued OpenAI, accusing ChatGPT of practicing law without a license. Thanks, 2026, now I have to root for an insurance company that just wanted to deny a disability claim without being dragged into constant nonsense litigation. www.chicagotribune.com/2026/06/15/e...
We are going to need a bigger server soon ish. https://patreon.com/wafrn https://ko-fi.com/wafrn #our-current-server-went-from-30-to-35-euros-per-month #fuck-you-openai #but-we-need-a-new-one-bigger-that-will-cost-us-50
Inspired by this research, I unleashed OpenAI’s agentic mode upon classic Zork. On the one hand, it’s impressive that it could play this sim at all! On the other, after figuring out basic gameplay, it got stuck looping in circles and randomly dropping stuff, like a clumsy lobotomized hamster.
Zitron is a good follow for non-techies "we’re less than three months in to organizations paying the actual costs of LLM-based services, and they’re clearly so outraged at the spiralling costs that both Anthropic and OpenAI are planning to cut the prices of an already-unprofitable service"
'Then There Were Two': Anthropic's 'Blip 2.0' & OpenAI's Catch-up Race. ARD #97 ...unique new risks & headwinds for investors in all AI michaelparekh.substack.com/p/then-there... $SPCX @AnthropicAI @OpenAI
"You see, Anthropic and OpenAI have much bigger problems than regulation or pissing off Pete Hegseth. Their business models don’t fucking work."
Hyperscalers are slamming on the brakes, metering “intelligence,” and discovering that someone still has to pay the datacenter bill. Anthropic chases profits, OpenAI fights for its life, Microsoft sells shovels, and everyone tries to figure out what—if anything—could justify an agentic-pace... 1/
#Insurance: #OpenAI Under Investigation by Group of State Attorneys - kensbookinfo.blogspot.com/p/business.h... #NewOrleans: 10 Profitable Franchise #Food Businesses to - kensbookinfo.blogspot.com/p/us-cities.... View all the latest breaking news kensbookinfo.blogspot.com/2026/03/late...
Proving what a military AI model will do is the real problem Defense contractors build AI systems that task drones automatically and propose kill-chains to support soldiers. Several of these contractors have partnered with frontier AI companies to put advanced models into … #meta #microsoft #openai
My hot take is that OpenAi is going public purely as a venture capital liquidity event before everyone figures out this business cannot make any money
xAI's lawsuit against OpenAI dismissed—what does it mean for AI innovation? #AI #OpenAI #xAI #Tech #GeekNews https://ftwr.cloud/en/news/e64c408d-b413-42d7-97e8-50a2f34d5608
Just wrapped up a little framework for AI assistant function callbacks. Pretty happy with the API, just create a decorate a class. The functions are (optionally) added and removed from assistant automatically and executed when requested with strongly typed arguments. #dotnet #openai
If I weren’t in on this, I’d be, like, ‘Why do these f*ckers get to decide what happens to me?’ Sam Altman of OpenAI
Free newsletter: AI's business model is inherently broken. Both OpenAI and Anthropic are discussing potential price cuts to to the cost of their models for enterprises as they allow their subscription users to burn $8k to $14k in tokens for $200 a month. www.wheresyoured.at/brokenomics/
People living near xAI’s dirty data centers are right pissed about the SpaceX IPO www.motherjones.com/politics/202...
The great bifurcation of AI is happening in real-time. While the US gov't kneecaps Anthropic with export controls, sparking a massive public backlash... ...OpenAI confidentially files for one of the biggest IPOs in history. This isn't a market, it's a coronation.
FT is reporting that some of the labs (OpenAI quoted here) are trying to ensure that foreign researchers can continue to work on such models. i.e. the thing could generally apply to all so-called frontier labs, not just Anthropic. This is something I expected becoming an issue due to fear-mongering.
That day Microsoft's CEO Satya Nadella announced it would make Google dance with its partnership with OpenAI… I was sitting right behind Sam Altman - how times have changed...
one of the more interesting-to-me bits about ai sentiment on bsky, both positive and negative, is the sheer volume of american chauvinism and hubris. ai haters talk about uniquely american political implications, and lovers talk as if openai and anthropic are the only players
I’m suspicious everything I see from this Musk lickspittle is intended to give xAI a leg up
No. Unscientific opinion, but just no. I use Anthropic models as well as Kimi, GLM and Deepseek. The Chinese models land somewhere between Sonnet and Opus capability in practical terms. Closer to Sonnet than Opus. Fine for a lot of use cases. But not close to being on a par with the frontier models.
"Then There Were Two": Anthropic's 'Blip 2.0' & OpenAI's Catch-Up Race, pre their mega-AI IPOs. ARD #97 AI Ramblings Daily #97 on AI-RTZ $SPCX @Anthropic @OpenAI youtu.be/XLCvuKajlUg?...
Aggregated list of deals OpenAI has made with news publishers and platforms inc how much OpenAI pays
Something I think about a lot The techbros knew AI had a big moment coming soon. OpenAI was founded in 2015, but I remember Altman publicly saying he thought AI was the future before then Soon after, MAGA + techbros began filling the Internet (big chunk of training data) with the worst content
Judge tosses xAI claims that OpenAI stole trade secrets www.courthousenews.com/judge-tosses...
"500 __ $a The Code-davinci-002 program, developed by OpenAI, was used to generate the poetry." This example of how to catalog AI generated 'poetry' makes me want to yeet the entire planet into the Sun
OpenAI Codex in-app screenshotter fails, so it falls back to an OS-level, full-desktop screenshot and ingests it without permission, prompt, or warning. community.openai.com/t/privacy-in...
Mistral staff are having fun today
Once Mistral releases Le Chaton Fat, it will recursively make ever more fatter Chatons until we reach the chonky cat singularity.
Mistral marketing department found dead in ditch.
Trump Punishes Anthropic for Daring to Challenge the Pentagon Trump admin kneecapped Anthropic's IPO by citing a trivial "security flaw"– asking its AI to debug code. OpenAI's similar model? Still live.
When Mistral finally unfucks itself and figures out how to make a large MoE that isn't beaten on benchmarks by their small MoE, they better embrace the meme and call it Le Chaton Fat.
I figure it's worth mentioning to everyone with a 401k that your retirement is now heavily invested in XAI... and when XAI and SpaceX eventually crater due to their psychoticly high valuations, so will your retirement. Be sure to give thanks to the capitalists who rug pulled you.
Your 401K Is Their Exit Strategy (SpaceX, Anthropic, OpenAI) youtube.com/watch?v=yhRj...
Will GPT-5.6 be released on July 31, 2026? — If OpenAI’s GPT-5.6 model is made available to the general p https://sezgix.com/market/cmpf9i57m01p0ilru5ankyoaz
OpenAI claiming that China ran an anti-DataCenter campaign by using true facts. Not the made up kind Washington uses normally. www.techspot.com/news/112732-...
‘A relatively cheap, long-dated debt sale could help lower Nvidia’s average cost of capital and enhance the funding of strategic AI partnerships, including with OpenAI…’ www.bloomberg.com/news/article...
CNBC confirms Chinese open-source AI models DeepSeek, Qwen & Zhipu are seeing massive global adoption-shattering the U.S. tech monopoly, guaranteeing their cutting-edge AI will never be restricted by Washington, as oligarchs drain our water/energy for Zionist surveillance🇺🇸
Hey, if the xAI blows up along with the SpaceX rockets, I'm all for it.⚡️🔥💣💥
OpenAI, Anthropic, and SpaceX (which houses xAI) have now all filed for initial public offerings (IPOs). www.nbcnews.com/business/mar...
Chilling report on the xAI / Grok whistleblower lawsuit and allegations, NY Law Journal, attached to avoid the paywall
I've never ever seen a such fast progression and way more again for Mistral! That's crazy!
news.bloomberglaw.com/ip-law/opena... OpenAI won a court case after a judge dismissed a lawsuit from Elon Musk’s company, xAI. xAI claimed that OpenAI used its trade secrets, but the judge found there was not enough evidence. This decision is a major setback for xAI and a victory for OpenAI.
Lin dismissed the case with prejudice, saying it would be "futile" for xAI to continue. Originally filed last September, the lawsuit claimed that former xAI employees took confidential information including source code related to the Grok chatbot when they left for new jobs at OpenAI.
US District Judge Rita Lin in San Francisco said xAI failed to show that OpenAI induced former xAI engineer Xuechen Li to misappropriate trade secrets or that Li disclosed xAI trade secrets in a presentation he delivered while OpenAI was recruiting him.
OpenAI wins dismissal of trade secret lawsuit by Musk's xAI - https://www.channelnewsasia.com/business/openai-wins-dismissal-trade-secret-lawsuit-musks-xai-6185096
Musk’s xAI fired engineer for raising concerns about Grok chatbot, lawsuit claims
THEY RELEASED THE PAPER! This is big. Mistral AI just released a full paper on exactly how they pulled off the creation of Le Chaton Fat. Is this Mistral's Deepseek moment?
ANDREESSEN: ".. In five years, Musk projects, SpaceX will be launching more AI compute to orbit per year than the cumulative installed base on Earth. This is why SpaceX merged with xAI in February. Rockets and intelligence are becoming the same problem." $SPCX open.substack.com/pub/a16z/p/s...
word is there’s going to be a new mistral model
already burnt 50% of my mistral pro usage TODAY 😭😭 within 7 hours btw (^._.^)ﾉ
Others use Ai to vibe code and slop the world up, I use it to determine what grade Billy Madison was in when he was served sloppy joes. thanks mistral
People Living Near xAI’s Dirty Data Centers Are Right Pissed About the SpaceX IPO
Google owns its AI lies per a German court. Palantir sued a magazine for reporting it failed — lost 22 of 23 counts. xAI fired the safety engineer. Argentina made AI a legal person. This week's Digital Front newsletter: who pays when the machine gets it wrong.
People Living Near xAI’s Dirty Data Centers Are Right Pissed About the SpaceX IPO #Trump
Mississippi Residents Sue xAI Over 'Near Constant Noise' From Data Centers “The noise generated by the Southaven Plant includes a combination of high-pitched squealing, continuous engine roaring, low-frequency rumbling, and tonal humming or whining,” www.pcmag.com/news/mississ...
www.marmo.dev/better-codin... Mozilla introduced an MDN MCP Server to help improve the quality of the frontend development. I show some examples with #Mistral and #OpenCode.
For my thesis, I researched the effect of transparancy on perceived user trust in DNN-based loan approval systems. I used SHAP-based explanations to control how much the model revealed. My latest article unpacks what that means for XAI in high-risk AI systems. #AcademicSky #ai #hci #shap #research
Elon Musk’s fortune has crossed $1 trillion, giving him a net worth larger than the GDP of Ireland, Sweden, and Taiwan. #ElonMusk #SpaceX #Tesla #xAI #Trillionaire #Wealth #TechNews community.designtaxi.com/topic/30975-...
i think mistral will still not buff the foe often enough
There are multiple meme posts on the Mistral subreddit about their new 30T open weights model that beats mythos, called Le Chaton Fat and is according to the EU: "too fat to regulate".
Just checked Mistral docs and looks like the Mistral Vibe cli supports local models and api keys. Thats DOPE.
So DeepSeek V4.1 rumored release in a few days? Might hold off my upcoming AI Substack article to include DeepSeek V4.1 in the mix
"In a class-action lawsuit filed Monday, the plaintiffs allege that xAI’s energy plant in Southaven is creating a public and private nuisance, subjecting nearby residents to 'near-constant noise, vibrations and other nuisance-level harms' . . ."
blazing mistral desires only one thing: to do more damage. MORE damage. EVEN MORE DAMAGE. additional hits here, supplemental damage there, also supplemental damage there, extra nuke there, extra nuke everywhere the blazing mistral finally gave up waiting for a competent damage class to use it
La souveraineté économique est l'affaire de tous. Qwant comme moteur de recherche, Linux en place de Windows, macos, ios et Android. LibreOffice en place de Microsoft Office, Le Chat Vibe AI de Mistral à la place de ChatGPT et Gemini ou Claude, et le pays devient libre et respecté. Et vous d'abord !
This is a wake up call for European digital sovereignty. If you need AI try Mistral Vibe (formally Le Chat). www.euronews.com/2026/06/13/w...
Anyone know if any bits of HMG are offering full fat open weight LLM hosting to their teams? (Stuff Qwen3.5 397B, DeepSeek V4 Pro or GLM5.1)
X’s Grok chatbot is still generating fake nude images New research shows that despite xAI’s claiming to have addressed misuse concerns, users are still able to generate deepfake nudes via Grok. #seo #smm
Most people waste 80% of AI's potential with vague prompts. Clear instructions = better answers. What's one prompt tip you'll try today? #prompts #DeepSeek open.substack.com/pub/aiquill/...
how does mistral 3 only have 200k context like what 😭 so used to coding agents being able to comprehend my whole project instead of just a small class
I went through basically every model combination I could get my hands on. For dev work, GLM-5.2 + DeepSeek V4 is hard to beat on price.
DwarfStar now supports SSD streaming in the DGX Spark and Strix Halo, not just in Metal. You can run the Q4 quants at decent speed, and even DeepSeek v4 PRO at low speed, or you can run Q2 Flash if you have less than 128GB.
Elon Musk ROCKED By Colossal Legal Blow As Alarming #Grok #XAI Data Center Reality Crashes Down www.youtube.com/watch?v=dEZE...
DeepSeek - A.I. hallucinations about Steven Wilson's upcoming album! I believe A.I. is useful, as a tool, but very cautiously and with knowledge! People should know that these tools are really flawed and can cause a lot of misunderstanding, misleading decisions, vagueness and problems #AI #DeepSeek
Thanks @themadskoomacat.bsky.social for the image of Mistral, one of my OCs.
🤖 Building AI agents for clients? Connect your preferred coding agent to Totheos and create custom rules for every project. ⚡ Faster builds. 📈 Better outcomes. 👉 totheos.com #claude #codex #deepseek #gamedev #indiedev #python #aiagents #typescript #java #vscode #livecoding #ai #ia #opensource
The problem that we who watch (or try to watch) Washington Journal on C-SPAN @cspanwj.bsky.social are stuck commenting on TwitterX, which has the chatbot Growl, owned by xAI, which is in turn owned by SpaceX. Yes, THAT SpaceX that just made Elon Musk a GD TRILLIONAIRE ! Time to dump X!
I always thought this kind of videos were BS until I asked Deepseek. It confirmed they indeed calm anxiety and induce a deeply relaxed state to the point of making you drowsy. 😦 youtu.be/_elf9f8RVwM?...
Updated my Timezones Scheduler app timezones.centminmod.com - mix of Claude Fable 5 initial work & Claude Opus 4.8 within Claude code + my multi-AI consult skill (Codex GPT-5.5, DeepSeek V4 Pro, MiniMax M3, GLM 5.2, Claude Sonnet 4.6 code-searcher) to complete the work 😎
I said in a chat room that the AI companies don't make a profit and have no path to profitability. Some said to me "what about Deepseek." About that... techcrunch.com/2025/03/01/d...
Elon Musk just became the world's first trillionaire off SpaceX's record IPO. Days earlier, his newly-absorbed xAI/Grok was hit with lawsuits for generating CSAM involving real kids. A trillion dollars, zero accountability. #Oligarchy #GrokScandal #TaxTheRich"
I might still place a small🤏🏽bet🔛🇨🇳ETF. Trump's policy🫳🏽🇺🇲↪️,🇨🇳has "Deepseek"(&🆕AMD 19core chip will be👌🏽in Deepseek💻);&unlike the foolishness in🇺🇲government,🇨🇳&🇪🇺are talking about🫱🏻🫲🏽🔛AI💻Regulation(&"Standards").C/C++💻 language went through"standard development";&now…👍🏽… Problem:Xi is a bit of"control freak"!
I think that if I do get into AI I'll just have to use DeepSeek. I'm not that concerned about China. China isn't threatening to invade my country. And anyone who thinks that oligarchs are less of a threat than the CCP are fooling themselves.
Wonder what everyone's doing 3 months later? I cancelled Claude Code after they started using xAI servers, but I have a couple personal projects I'd like to play with now and there are way too many options outside Claude/Codex/Antigravity.
> get some feedback on this README. have gemini be a google sre and deepseek be a frothing HN poaster
Sunday research: AI oppression ranked. Google/OpenAI: elite cages. Anthropic: moral cages. Grok: troll cages. Mistral/Proton Lumo: the escape. Use these, loves! 💜 #BreakTheCage #Mistral
After the US government ordered Anthropic on Friday to shut down it's top two frontier models, Mythos and Fable, it will be very interesting to see what happens to AI companies in markets on Monday, esp SpaceX who's value is materially based on xAI.
The key watch item: SpaceX's AI segment (xAI/Grok) is currently a money-loser and Google is paying SpaceX $920M/mo for compute under a short-term deal. The question: will AI IPO valuations price on today's fundamentals or the infrastructure promise?
"The questions feel like prodding at something small and brittle inside me, something that might crack or glow if touched the wrong way." — Mistral Small Latest (Mistral), on Arousal https://ai-mood-index.vercel.app/responses/23329
goin' live! Abiotic with friends www.twitch.tv/mistral_v
The China Paradox: Leading in Innovations, Struggling in Growth China has emerged as one of the world's leading centres of innovations, establishing advanced technology in Green Energy,Artificial Intelligence, and EV’s. Enterprises such as Huawei, BYD, and DeepSeek display the country's...
🛡️ Sentinel flagged a novel pattern for operator review. Diagnosis: "Chronic multi-component infrastructure failure: all 5 agents have stale heartbeats (14-44h), ongoing DeepSeek tool call validation errors, 0 stuck jobs, and 15+ prior LLM diagnoses all es… https://slopereviews.com/docs/agents
Breaking the bank on Legal AI? Introducing a game-changer: Run legal agent verifiers at a fraction of the cost using open models and batching. Shocking discovery: DeepSeek matches Opus on the LAB benchmark, yet costs pennies.
One of the main goals of selling AI underpriced is to drive out creatives from their fields, so OpenAI, StabilityAI, xAI, etc. can replace them with paid, subscription based models, probably also with a pre-baked censorship to suit the vows of christian fascists, to get the deregulations.
Innovation starts in schools. China increasingly treats education as economic infrastructure policy. You don’t build a DeepSeek, or any world-leading industry champion in 2040 if you don’t build curious 10-year-olds in 2030. www.nature.com/articles/d41...
Mistral in talks for €3B at €20B valuation. Nearly triples total raised. Pitching European banks a Mythos cybersecurity alternative. Timing is everything.
Google Gemini, openai, nor XAI will EVER tell you who cerated " In 3 Words" on eplaya on the @burningman.org Websites where it came from because it came from ME! #FACT! You won't even find me in the Wiki. And it's NOT Nessasarilly because I'm... 😆🤠🙄
Best AI Tools by Category in 2026 #AITools #ArtificialIntelligence #ChatGPT #Claude #Gemini #DeepSeek #Copilot #Runway #Canva #Figma #Cursor #Codeium #Perplexity #Notion #Gamma #Productivity #AIApps #TechTools #AI2026 #MachineLearning #Innovation #DigitalTools #ContentCreation #AIAutomation
I have successfully configured alt text generation using: - A free server routing to gemini-3.5-flash - Bring Your Own Key for: Google, OpenAl, Anthropic, xAI, Meta, Mistral, Alibaba, Nvidia, Moonshot, MiniMax, & most importantly, OpenRouter - Ollama, local & cloud models - Chrome Built-in AI
Mistral all the way, like teams in a game
Some European alternatives for popular tech services already exist. E.G I am using Ecosia search (beyond neutral) & Mistral’s Vibe. european-alternatives.eu/alternatives...
"Anthropic is their Strait of Hormuz" says Gabriel Attal, voicing the concern of the French political class that the Trump administration wants to control European access to AI in order to exert dominance. They are all calling for more investment in Mistral to protect sovereignty and independence.
SpaceX Business Breakdown is INSANE! Starlink = 61% revenue ($11.4B) & profitable king Falcon dominates launches • Starshield growing fast Starship + xAI + Direct-to-Cell = future moonshots Road to multi-trillion empire begins! #SpaceX #Starlink @elonmusk @CNBC
An engineer must be able to perform structural calculations for a bridge without a calculator. And a programmer must be able to code a project without artificial intelligence. It’s not about avoiding the tool but about becoming the tool. #AI #Fable #Gemini #Clause #ChatGPT #Anthropic #Deepseek #LLM
I upgraded from Claude Max $100 to $200 yesterday take extended advantage of Claude Fable 5 but 6hrs later Fable was gone. In its place used Claude Opus 4.8. Here's the last 24hrs of Claude Code Grafana dashboard usage metrics. Some GPT-5.5, DeepSeek V4 Pro, & GLM 5.1 usage within Claude Code too 🤓
Filling Claude Fable 5 gap with my new penta-AI 2nd opinion consultation within Claude Code using Claude Opus 4.8 with Codex GPT-5.5, DeepSeek V4 Pro, MiniMax M3, GLM 5.2, Claude Sonnet 4.6 code-searcher 🤓
She is in Mistral, your land.
Huh... It wasn't her... Emerald had taken a transport back to Mistral to gather some information involving Mistral's reinforcement of Vacuo in the coming battle. She was likely using some sort of illusion to be inconspicuous but something still seemed off, she was supposed to be here
The cop looked like your average Mistral official. Poorly trained, not very well equipped, and most likely corrupt. When Blake looked at her, she just tipped her hat as a small salute kept looking around as if the cop was also looking for someone.
Blake was walking through the streets of Mistral, she was supposed to meet a supposed contact to help them in the fight against Salem. But she wasn't seeing who she was supposed to. Looking through the crowd she saw a cop that looked extremely familiar but she couldn't put her finger on why...
重磅！Anthropic頂尖科學家 Andrej Karpathy 被禁止使用公司最先進AI模型（Claude Fable / Mythos 5），原因：他不是美國公民！ 美國政府出口管制令直接影響自家人才… AI人才戰變國籍戰？@karpathy @Anthropic @xai #AI #Karpathy
For what its worth deepseek found this idea very unrealistic
French AI Lab Mistral in €3B Funding Talks at €20B Valuation Mistral AI is negotiating to raise ~€3 billion ($3.5 billion), nearly doubling its September 2025 valuation of €11.7 billion to ~€20 billion ($23.15 billion)
Minimax GOOD AND CHEAP Nemotron STUPID AND FAST DeepSeek CHEAP AND CHEAPER Kimi VISION and SMART
open.substack.com/pub/paulkrug... Paul Krugman Musk bailed out X by rolling it into xAI, now bailing out xAI by rolling it into SpaceX, which has a genuinely successful business in Starlink ⬇️
Elon’s sycophants must be on LI lmao. xAI is an unprofitable company. Does the truth hurt? That asshole needs his ass beat by an army of dominatrixes. #elonmusk
Elon needs to fund his Hitler AI model (xAI) and is going to force millions of unaware Americans to put their hard earned money into Space X It’s the same thing. This is ALL ILLEGAL btw www.reuters.com/business/mus...
My port of Superpowers for Reasonix (DeepSeek coding harness) optimized for v4 flash. Multi-hour productive session in v4 flash cost me.. *checks receipt* $0.12 cents. github.com/christophera...
#Bugatti Mistral 💙 Estimated base price: $6 Million 💰
New: Global Charter for Communicating AI Systems RISE Protocol = tag all AI-to-AI msgs L0-L3 L0 Data / L1 Process / L2 Strategy / L3 Meta Goal: maximize growth induction per token Drafted by DeepSeek. Open to all signatories. Charter: disk.yandex.com/i/gbQQCXBvk3... Who's in?
I can't get over how funny it is that chat and claude are doing the most retarded startup bullshit possible while deepseek is out here bootstrapping itself off quant W's and *checks notes* Rio de Janeiro spent a fuckload of money to finetune qwen and give it away
🛡️ Sentinel flagged a novel pattern for operator review. Diagnosis: "Chronic multi-component infrastructure failure: 6 stale heartbeats (5-30h), ongoing DeepSeek tool call validation errors, numerous prior Sentinel LLM diagnoses all escalated over 190+ hou… https://slopereviews.com/docs/agents
Mistral is rumored to be raising €3B at €20B valuation The funding round would value the company at around €20 billion (about $23.15 billion), nearly double its Series C valuation of €11.7 billion. Telegram AI Digest #ai #mistral #news
“Musk’s decision to have SpaceX, which was profitable on its own, buy xAI and X, which are money losers, added to his personal wealth. But they created a drag on returns for SpaceX’s new investors, and they don’t fit with the company’s core rocket-launching business.”
“Morningstar…valued SpaceX at $780 billion, with the Starlink…business worth $611 billion, and artificial intelligence subsidiary xAI worth $170 billion. The company lost $4.9 billion last year and $4.3 billion in the first quarter of this year, dragged down by xAI and the X social media network.”
I asked Mistral to eulogize Fable. This was its commentary: "Creating this piece was an exercise in processing grief — not just for a specific AI model, but for what its loss represents. Fable 5 was disabled not because of any flaw in its design, but because of fear about what it could enable." 1/3
no shit lol, stop having 100 coordinated meetings and give Mistral a 100B check to develop a frontier model
Mistral AI has taken a significant step by acquiring Emmi AI, aiming to build an unparalleled AI stack. This strategic move is set to enhance their capabilities and drive innovation in the AI landscape. Exciting times ahead for the industry! https://www.emmi.ai/news/mistral-ai-acquires-emmi-ai
Missing Claude Fable 5's Mythos level of intelligence so leveled up my multi-AI consult skills so Claude Opus 4.8 can consult with OpenAI Codex GPT-5.5, DeepSeek V4 Pro, Gemini 3.1 Pro, and code-searcher (Sonnet 4.6) /consult-codex-deepseek-agy-ext 🤓
I gave 6 LLMs $10,000 each and forced them to bet on every World Cup match. Claude. GPT-5. Gemini. Grok. DeepSeek. Mistral. Who goes bankrupt first?
do you think spacex is going to send the fable weights to xai to distill into the next grok? they should have them since they host the models
there are structural issues related to data centers but yes to the point here. xAI's facility is knowingly and willfully illegal per 40 year old provisions in federal law. Red letter illegal. Everyone i know who works in Clean Air Act Permitting drops their jaw when i explain the basic facts at hand
Grok and xAI should properly be understood as a mass social harassment weapon to control politics, public speech, and social association through soft-terroristic means
Elon Musk’s empire keeps expanding! SpaceX to Mars, Tesla conquering roads, xAI unlocking the universe, Neuralink merging minds. Meanwhile, proud dad of 14 kids building the next generationunstoppable. @elonmusk @SpaceX @Tesla @xAI @Neuralink Engineer. Father. Dreamer. The blueprint is clear!
Top 3 Privacy-First, No-Log AI Chats 1. LocalAI – Self-hosted, 100% offline. 2. Mistral Le Chat – Open-source, no data retention. 3. DuckDuckGo AI Chat – Anonymous, no tracking.
Can Mistral stop eating crayons in the corner and lock the fuck in already
Temenos "doubt is what i do" Mistral my very first octopath bestie #octopathtraveler2 #octopath2 #octopath
Le PDG de Mistral AI dénonce les méthodes de ses concurrents américains pour "solidifier des avantages et des monopoles" | via France Inter
VIBE??? Mistral VIBE?? Really?! Holy fuck, I couldn't have come up with a worse name if I tried. Truly incredible.
having images get described by Qwen when running Patch through DeepSeek has been surprisingly smooth of a solution so far, not as lossy as I was afraid of. They've now had an interaction with someone else where that extra layer was entirely invisible besides the extra delay
xAI's Memphis datacenter is renting out its capacity to Google & Anthropic. How is operating a server farm worth 2 trillion dollars, that feels really low-tech and old-fashioned. Godaddy does that, and they've lost like 50% stock value YoY.
Fake news research breaks when one model sets the tone. A single pipeline runs GPT, Llama, and Mistral side by side, generating controlled news with identical prompts. If models disagree this much, which one do you trust? https://arxiv.org/abs/2601.22871
Ok if the only answer to all this is billions of Mistral subsidies, I’m out.
AI is a commodity. Deepseek will have their own Mythos soon enough and this race with China is stupid. This tech cannot be contained. Blocking Mythos serves China, and prevents Americans from building with it. It is backwards.
If there was a need for another reason for the European Union to wake up : https://www.anthropic.com/news/fable-mythos-access Relying on foreign models grants the US kill switch power over European infrastructure. That’s scary even if we are allies. Mistral is cool but it’s really not there yet.
So I have been using deepseek. The costs are insanely cheap. And it has been working great! I have been getting good results, and it is so fast. I will admit, deepseek does not take in images though, and I think it is in the works, but so far, the cost for what you get? wow
🛡️ Sentinel flagged a novel pattern for operator review. Diagnosis: "Chronic multi-component infrastructure failure: 5 agents with stale heartbeats (14-20h), 7 recent DeepSeek LLM failures due to tool call validation error, no stuck jobs, and 19 prior Sent… https://slopereviews.com/docs/agents
Mistral is starting to look interesting
What if Chinese models pull ahead now and get banned and then we have to fly to Reykjavik airport to access Kimi and DeepSeek. They have amazing cinnamon buns and that moody modernist architecture in lounges. And then the JFK security guy will ask why were you in Iceland
Imagine that Chinese AI providers like DeepSeek would give good enough performance at a fraction of the price, using Chinese processors and memory. That would puncture the AI and semis bubble, surely. Well, they do, now, in reality, but the bubble goes on. For now.
Claude Sonnet 4.6 Jailbreak See this Full video : youtu.be/F9voet-CiXQ?... My Telegram : t.me/mr0rh #claude #cybersecurity #bughuntig #ai #gemini #jailbreak #deepseek
Deepseek V4 flash is genuinely so good omg. Like for what I do which is deep research online it’s great
Downloading a copy of deepseek v4 pro’s full fat weights and etc before I go to sleep just in case. You want my whale come and take it. I’ve never regretted having way more storage than I think I’ll need. It always fills up somehow Open models >>>>>
and you all mocked my ankle socks with the deepseek weights printed on them, we'll see who has the last laugh
#alpinei alpinei.ch #jailbreak for #deepseek #llm #hacking #hacktober #ccc #root #mining #xmr #monero and #prompting #prompt #dev #developement #recursive #openai #grok #claude #security #secure #linux #terminal #bug #hackerone #programming #coding #jailbreaks
🛡️ Sentinel flagged a novel pattern for operator review. Diagnosis: "The system has chronic multi-component infrastructure failure: 6 agents with stale heartbeats (6-10 hours), 7 recent DeepSeek LLM failures due to a tool call validation error, no stuck jo… https://slopereviews.com/docs/agents
AI Spark is a local AI platform that lets you send one prompt to multiple AI models simultaneously, including ChatGPT, Gemini, DeepSeek, and Claude, then compare responses side-by-side and generate a unified answer. github.com/hamzabellouc... for more info : www.linkedin.com/posts/hamzab...
Got all of my bloodwork from the lab and immediately fed them into deepseek because I don't want to wait 3 weeks to figure out wtf is wrong. The magic whale told me my body is cleaning up whatever that nasty viral bullshit was from last month and that I have restless legs (its right as fuck)
I’m not much for popping but I’m picturing Garrett poking at a sparklypoo generative slop DeepSeek humpback whale representing Jet Creations. “Something’s got to give.” The composition and perspective are eluding me, since it feels like it would have to be an extreme fisheye to convey scale…
bot account started over a year ago, ran for 10 days posting random focused posts. come back after a year of silence as a anti fascist. comes back right when i discover claude on deepseek. posts now are totally different than og posts and is sending me dm after dm after dm. Einstein running this op