Bran

200 posts

Bran banner
Bran

Bran

@Bran_Fi

Pursue your curiosity with unwavering passion and with the intensity of a dying star.

Sumali Şubat 2024
952 Sinusundan18 Mga Tagasunod
Bran
Bran@Bran_Fi·
@DavidSacks Definitely need better PR people that's for sure. WH initially blaming it on the "Fix this code" prompt which still required manual reverse engineering/actually writing code though was equally bad communication. That's saying "any llm that can find bugs in code" should be banned
English
0
0
1
458
David Sacks
David Sacks@DavidSacks·
This brilliant lesson on communication by @rabois is why the AI leaders are failing. It’s not sufficient just to “speak your truth.” You have to communicate in a way that elucidates your audience. Convincing the public that your company is a menace obviously fails that test.
English
70
52
870
95.5K
Bran
Bran@Bran_Fi·
@signulll The so called jail break was "Fix my code" everything else had to be done manually by someone with the prerequisite knowledge to reverse engineer the fix into a script/exploit because Fable refused to do it directly or automate that process like it was suppose to. ¯\_(ツ)_/¯
English
0
0
0
219
signüll
signüll@signulll·
let me get this straight. anthropic launches fable. it took few days for the u.s. government to force it offline. all cuz amazon discovered a jailbreak which appears to be a standard cyber security workflow that asks the model to fix vulnerable code. is this a giant cyber security vulnerability? well anthropic doesn’t think so. you can do this with gpt 5.5. so why single fable out for this specifically? if you do, then should every serious model be export controlled the moment it can do useful security work? didn’t matter cuz the admin dropped an export control order anyway. sacks now says we told dario to fix it which is interesting because what exactly is this fix? how do you fix this? more specifically what exactly is the regulation? are you regulating the model itself? a capability? a prompt pattern? some cybersecurity workflow? seems to me that anthropic would have a lot to lose if this model actually caused serious widespread security vulnerabilities in the first place.
English
82
20
827
82.9K
Michal Malewicz
Michal Malewicz@michalmalewicz·
Tidy mac, tidy mind. Or something.
Michal Malewicz tweet media
English
2
0
18
1K
Alex Imas
Alex Imas@alexolegimas·
The original with my UChicago colleague @Ada_Palmer was perhaps my favorite podcast episode ever. Cannot wait to listen to the sequel.
Dwarkesh Patel@dwarkesh_sp

Had @Ada_Palmer back on – this time to talk about Machiavelli, perhaps the most misunderstood thinker of all time. Machiavelli cut his teeth as a high-level diplomat for Florence, a position from which he got to closely observe the most important rulers in Europe at the time, including the ones who were on the path to destroying his dearly beloved Florence. In 1513 the Medici retook control of Florence and, wrongly suspecting Machiavelli of participating in a coup attempt, fired, tortured, and exiled him. Machiavelli could have fled his exile and worked for any number of different principalities that would have been eager to make use of his talents. Instead, he decided to rot in the countryside and compile his career's lessons about power, politics, and human nature into a book he dedicated to the very man whose new regime had tortured and exiled him, Lorenzo di Piero de' Medici. But at least the Medici were in a position to use his insights to defend Florence. Machiavelli the patriot did not want any other hands to touch this book, because those hands, armed with these lessons, might pose an existential danger to Florence. The closest modern analogy, at least as Machiavelli would have seen it, would be Szilard's letter warning FDR about the possibility of a nuclear fission bomb. What were those insights? And how were they inspired by Machiavelli's dangerous diplomatic missions all across Europe, and his extensive reading of antiquity? Watch this episode with Ada Palmer to find out! 00:00:00 – How Florence bargained with Cesare Borgia for survival 00:15:08 – Machiavelli’s analytical innovations 00:23:58 – Why popes became warlords 00:36:13 – Why the common people demanded nepotism 00:47:57 – Cesare Borgia brought terror to rulers and justice to the people 00:57:55 – Art as a proxy for war 01:06:41 – Florence, a city famous in hell 01:15:57 – The Prince was a job application to Machiavelli’s torturers 01:41:39 – During the Renaissance, original ideas had to be couched in antiquity 01:50:44 – Why copyright began with the Inquisition 02:02:12 – Machiavelli wasn’t Machiavellian Look up Dwarkesh Podcast on Spotify, Apple Podcasts, YouTube, etc.

English
4
5
136
14.1K
Bran
Bran@Bran_Fi·
@DiligentDenizen @RepThomasMassie @RepRoKhanna FY2027 National Defense Authorization Act includes a proposal known as the United States-Israel Defense Technology Cooperation Initiative, which seeks to synchronize joint research, weapons coproduction, and integration of advanced technologies like AI/Cyber defense . Still bad.
English
1
0
2
112
Diligent Denizen 🇺🇸
Diligent Denizen 🇺🇸@DiligentDenizen·
‼️🇺🇸: AMERICANS CALL YOUR REPRESENTATIVES KEEP THE USA FREE CALL THESE 13 👇🏻 congressmen and tell to support @RepThomasMassie & @RepRoKhanna amendment removing the Israel-US miltary tech merger to the NDAA listed top down of least $ from Israel lobbies: Ralph Norman (R-SC): (202) 225-5501 Chip Roy (R-TX): (202) 225-4236 Erin Houchin (R-IN): (202) 225-5315 Morgan Griffith (R-VA): (202) 225-3861 Austin Scott (R-GA): (202) 225-6531 Michelle Fischbach (R-MN): (202) 225-2165 Mary Gay Scanlon (D-PA): (202) 225-2011 Jim McGovern (D-MA): (202) 225-6101 Teresa Leger Fernández (D-NM): (202) 225-6190 Nick Langworthy (R-NY): (202) 225-3161 Virginia Foxx (R-NC): (202) 225-2071 Joe Neguse (D-CO): (202) 225-2161 Brian Jack (R-GA): (202) 225-5901
English
25
467
831
20K
Bran
Bran@Bran_Fi·
@elonmusk Looks like the U.S. Government is wanting to do this with all AI/LLMs. Full KYC with multiple forms of ID+Full data retention input/output, thats what this will turn into if they get their way with this whole Anthropic fiasco.
English
0
0
2
123
Bran
Bran@Bran_Fi·
@neuralease @emollick Open source/weights won't be catching up to 10+ Trillion parameter models with current LLM architecture. Would need something new. You wouldn't be able to even serve it, even most businesses couldn't even on-prem it, you would need a private data center and deep pockets.
English
1
0
0
8
Neuralease
Neuralease@neuralease·
@emollick This would be a horrendous future, democracy will be a fairytale. Unless Open Source catches up.
English
1
0
1
70
Ethan Mollick
Ethan Mollick@emollick·
If AGI is achievable & labs can be banned from using a model internally ONLY if they release the model publicly, the Big Three labs may decide it is better to capture all the value from AGI themselves by expansion & acquisition. Sharing AI access with other firms triggers risk.
English
76
77
1.2K
67.4K
Bran
Bran@Bran_Fi·
@theo From what I've heard they don't like the fact that AI/LLMs can find any kind of bugs in code period. Even that is being seen as a threat I don't see this being resolved anytime soon. Other labs will probably be called in. There is too much stupid in the WH right now.
English
0
0
0
1.8K
Theo - t3.gg
Theo - t3.gg@theo·
It's kind of wild that Fable still isn't back. Honestly thought this would be resolved quicker 🙃
English
224
40
3.6K
172.1K
Bran
Bran@Bran_Fi·
@signulll It's kind of hard to argue about it being disease like at times. Brain rot is a thing. 😐 Sounds more like they just need to learn to filter the right things though and build immunity. Idiots will be idiots anywhere they are, it's just easier to see on social media.
English
0
0
1
745
signüll
signüll@signulll·
it’s fascinating how much of anthropic’s posture seems shaped by a genuine revulsion toward social media. dario & daniela talk about the social media landscape like a disease more than a product category (latest interview). after this you can see anthropic’s macro decisions start to make a lot more sense incl. the enterprise focus, the anti engagement loop ethos, among lots of other things highlighted in the constitution. a kind of modern anti feed doctrine if you will.
English
52
15
735
46.7K
Bran
Bran@Bran_Fi·
@elder_plinius Government is already upset and wants to ban any AI that can find any kind of bugs in code based on their current talking points. Yikes. Massive high level incompetency taken to the extreme in the WH.
English
0
0
2
645
Bran
Bran@Bran_Fi·
@CalimanuLoredan @elder_plinius The kind of emergent intelligence you see in these 8-10 Trillion + param models is not going to be running local/on-prem with current LLM architecture. A dream within a dream. Some wildly new system would have to emerge to achieve such a feat.
English
0
0
0
56
Calimanu Loredan
Calimanu Loredan@CalimanuLoredan·
@elder_plinius You believe we can run open source models locally on consumer hardware at Fable level capabilities in 3-6 months?
English
20
1
69
13.4K
Bran
Bran@Bran_Fi·
@jxnlco How to make Codex with 5.5 produce output as good a Clade Code with Fable 5/Mythos. 😂 I jest. Of course then the government would just ban consumer AI at that point lol. They already want to ban AI from finding any bugs in code.
English
1
0
0
5.3K
jason
jason@jxnlco·
so close to 90k followers... what codex articles do i need to write to get there...
English
105
10
430
545.6K
Bran
Bran@Bran_Fi·
@MTSlive Axios said someone "claimed" Anthropic failed to honor a recent cyber executive order. Not that it was actually determined as fact, details were not shared of any kind what-so-ever.
English
0
0
1
56
MTS
MTS@MTSlive·
SITUATION EXPLAINED: The Trump administration just put an export ban on Claude Fable. Here's what we know: • Axios reporting: Anthropic "failed to honor a recent cyber executive order" • Andy Jassy, CEO of Amazon, a major Anthropic investor and cloud partner, personally called Treasury Secretary Bessent to raise concerns that Mythos and Fable could be jailbroken • Users reported being blocked for basic biology questions • Anthropic announced Fable months in advance specifically to implement safeguards, now being criticized for not being safetyist enough @gbrl_dick: "The same people who were criticizing Anthropic for releasing the model with these quite clumsy, aggressive safeguards are now criticizing the admin for this decision."
MTS@MTSlive

SITUATION UPDATE: Anthropic confirms it is disabling Mythos and Fable 5 for all customers to comply with a US government export control directive. The company says it believes the order is a misunderstanding and is working to restore access.

English
5
1
32
8.5K
Bran
Bran@Bran_Fi·
@Keir_Starmer This will not protect children or enrich their lives. It'll just inflate egos of adults thinking that is what they are doing. It will isolate them even further. They will be ignorant on the complexities of the world stage where everything happens online. This is cultural suicide
English
0
1
1
5
Keir Starmer
Keir Starmer@Keir_Starmer·
We are banning social media access for under 16s. These days kids must find their feet in a world where technology intrudes into every area of their life. I just can’t let that go on anymore. So we’re giving children their childhoods back.
English
38.8K
5.9K
35.2K
14.2M
Bran
Bran@Bran_Fi·
@krispuckett @perplexity_ai @BlitzyAI has been using a similar approach in their product since 2023. They have some several other parts to it though that constitute as their secret sauce from what I've heard.
English
0
0
0
33
Wawa Sensei
Wawa Sensei@wawasensei·
Asked Fable to code Fable lite with Threejs/WebGPU One prompt result 👇
English
6
1
63
6.9K
Bran
Bran@Bran_Fi·
@wawasensei @suno @ElevenLabs @dangreenheck Next make it pvp multiplayer keep the chickens in have them swarm the closest targets and have to defend against them and aggressive players 🤔 add a movement ability so you can run away after training them with chickens 😵 Last man standing rules.
English
1
0
1
41
Wawa Sensei
Wawa Sensei@wawasensei·
@suno @ElevenLabs @dangreenheck Before Fable went down, I was able to add a few updates: - Motion blur - Health/Mana potions - Kamikaze Chicken 💥 - Mobile version I'll put it on hold, praying it comes back soon 🙏
English
4
1
22
3.7K
Bran
Bran@Bran_Fi·
@TomDavidsonX Government is basically saying their feelings and intuition on the matter is more important than facts, objective reality, and long term consequences. Seeing this happen in real time is painful to watch, such high level incompetency. Same as it ever was.
English
0
0
1
329
Tom Davidson
Tom Davidson@TomDavidsonX·
Seeing some ppl interpreting the Fable shutdown as positive news on govt intervention But i don't think we'll execute a pause well*, or use the pause effectively, if the govt has NO idea how ai tech works and just wades in arbitrarily without consulting experts it's def evidence for "govt might do lots of random extreme stuff" -- and that's prob good if you're p(doom) is v high and you want to fuck shit up and shake up the game board. but still, if you think alignment is v hard then ultimately we need sane ppl in charge to handle it *executing a pause well is hard! You need to figure out effective verification, strike a deal with China, avoid extreme power concentration, catch rogue projects, maintain safety standards even as many unsafe projects catch up to the frontier
English
21
22
303
39.7K
Ethan Mollick
Ethan Mollick@emollick·
A plus of Fable being down is the number of times the word “toast” appears in Claude Code has dropped dramatically. That model loved/loves software development and UX jargon more than any other model I have used.
English
20
2
170
20.8K