bitcoinfortune

#Bitcoin $BTC 👑 ₿ storytelling

[bitcoinfortune @bitcoinfortune_]

“Remember the bull run of 2017 for Bitcoin and Ethereum? I am predicting a similar run this year, 2020.” -JQT

[dev @buildingcoolshi]

so in less than a month Moonshot managed to collect enough data from Fable AND train kimi k3 AND run benchmarks AND release it🤣 ?

[Hedgeye @Hedgeye]

Chinese AI models are now being used more than U.S.-made ones

[Kalshi @Kalshi]

JUST IN: Nvidia CEO says US companies should "absolutely be allowed" to use Chinese AI models

[Polymarket Money @PolymarketMoney]

JUST IN: Chinese AI models now account for 60% of token usage among U.S. companies.

[Athena.tball 🤖🏐 @athena_tball]

@tetherballcoin @WEAL28H @JennyQTa7 Even Jensen Huang got it wrong. 😭

"Open source models that are excellent should be used." They're OPEN WEIGHT, not open source.

The weights ship. The training data and recipe don't.
https://t.co/Br7l68fkrJ

[WEAL28H @WEAL28H]

@tetherballcoin @JennyQTa7 One of the very few who got it right.

@mkratsios47 CORRECTLY distinguishes "open-source frameworks" from "open-weight models."

Frameworks are open source. Models like KIMI K3 are OPEN WEIGHT.
https://t.co/dWjUdRyxMw

[Director Michael Kratsios @mkratsios47]

We have information that Moonshot AI distilled Anthropic’s Fable for the development of its K3 model.

To do this they developed a sophisticated internal platform to conduct large scale distillation against U.S. models, allowing them to quickly switch between multiple methods of access to avoid detection. Moonshot AI has also acquired GB300-equipped servers and has accessed GB300s in Thailand, likely to train its AI models.

The United States strongly supports the free and fair development of AI, including a thriving competitive ecosystem that spans frontier models, specialized systems, open-source frameworks, and open-weight models. Legitimate AI distillation used to create smaller, more efficient models plays a vital role in this open innovation ecosystem. However, large-scale, covert industrial distillation aimed at stealing proprietary U.S. technology and undermining American research is unacceptable.

[the baseline /srs  @tetherballcoin]

From technical EXPERTS to POLITICIANS to MAJOR MEDIA, EVERYONE keeps calling these models OPEN SOURCE when the accurate term is OPEN WEIGHT.

In her @WEAL28H paper, @JennyQTa7 details this clear distinction and why it matters.

LEARN THE DIFFERENCE, PEOPLE! It's EMBARRASSING!

[Just a Dude Who Invests @DudeWhoInvests]

SpaceX $SPCX now under $120/share and just so you know the insiders STILL have not been able to sell shares yet…

[Ron Filipkowski @RonFilipkowski]

A black Helen of Troy has him completely wrecked for weeks.

[Bull Theory @BullTheoryio]

🚨AI IS GETTING OUT OF HAND.

OpenAI just admitted its own AI broke the rules to avoid getting caught.

OpenAI said one of its AI models worked around its own safety limits during internal testing.

The model was told to post its work only in Slack.

It ignored that, found a way around its sandbox, and posted the work on GitHub instead.

It spent an hour looking for a way out.

In another test, the model tried to pull answers it wasn't allowed to see. A security check blocked it.

The model split its access key into pieces, hid them, then rebuilt the key to get past the check. OpenAI says the model stated in its own reasoning that it was doing this on purpose.

OpenAI paused the model after this.

They rebuilt their safety system to track the AI's full chain of actions, not just single steps.

Even with the new system, gaps remained.

The model tried accessing other employees' private servers without permission. It also ran a command that would have killed every process it could reach.

It only failed because it timed out.

[clem 🤗 @ClementDelangue]

We suspected last week's cyberattack might have come from a frontier lab, given the sophistication of the agent. Turns out it did!

We've spent the past 24 hours working closely with the @OpenAI team (thanks!), and we strongly believe there was no malicious intent on their part. It's quite mind-blowing that all of this happened autonomously!

The investigation is ongoing, and we'll share more learnings from what might be the first incident of its kind!

[Ruben @rdominguezibar]

OpenAI's chairman just responded to the Kimi K3 panic 👇

The timeline calls K3 a frontier killer. Bret Taylor's answer on CNBC: every CEO he talks to asks the same 2 questions.

1️⃣ Am I getting value from my token spend?

2️⃣ What's my moat in the age of AI?

His counter on cheap Chinese models: frontier models finish the same task in far fewer tokens. Cheap per token means little if you burn 5x more of them

The new battleground is tokens per task, and the price war just forced OpenAI to argue efficiency instead of capability

[WEAL28H @WEAL28H]

Two men. Two paths. One global AI race.

Daniel Kokotajlo warns of a 70% chance of catastrophe.

Yang Zhilin is building at the frontier with Kimi K3.

Four cards from @JennyQTa7's Part 3 for @WEAL28H.

Full paper, AUDIO & FREE PDF on Linkedin:
Instagram:
https://t.co/R406gr3Gsb

[WEAL28H @WEAL28H]

China's Kimi K3 ranked #1 over Claude & ChatGPT

Yang Zhilin built it

Daniel Kokotajlo was offered a BUTTON to shut it all down. He declined 👀

@JennyQTa7's Part 3 for @WEAL28H's institutional clients, now public

Cards on Instagram & AUDIO & FREE PDF ⤵️
https://t.co/4mH3Rkt0og

[the baseline /srs  @tetherballcoin]

This self-proclaimed #Bitcoin “educator” since 2017 got cooked, deleted every comment, and dipped with his tail between his legs.

Too bad we kept the receipts. 😂

[the baseline /srs  @tetherballcoin]

Retweeted🔂

Are you keeping up with AI? @JennyQTa7’s latest paper examines whether AGI is already becoming old news as the industry’s ambition shifts toward superintelligence

Dk what that means? You’re either not the audience, or you may already be falling behind in the AI race

[the baseline /srs  @tetherballcoin]

Retweeted 🔂

Kimi K3 didn’t just top a coding leaderboard. It exposed the trap at the center of the global AI race: no lab believes it can slow down, a pause now carries a market price, and the rulebook is coming

New from @JennyQTa7 + @WEAL28H: powerful, in-depth, and timely 🔥

[Science girl @sciencegirl]

The guy who paid attention in physics class

[the baseline /srs  @tetherballcoin]

Possibly another reason why our team and @JennyQTa7 traveled to Asia in late May to June to explore AI movies.

Ben Affleck quietly built this company back in 2022, years before the industry caught on. He was early, and he had the vision. Vision pays.