iWyzer (@averagewyzer) header
iWyzer (@averagewyzer) profile picture

Random dude that loves tech 🫡

Web 3.0
Joined October 2021
Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)Photo by iWyzer (@averagewyzer)
Hey YOU 🫵 ! Have you ever wondered If you had the "right" education ? Or if you had the right path to do what you want ? Does it define your intellect ? Let's deep dive into some education questions. Read this 🧵👇 (1/16)
9
8
1
49
4,531
iWyzer retweeted
SITUATION DETECTED: Hugging Face co-founder Thomas Wolf, xAI co-founder Igor Babuschkin, MIRI President Nate Soares, and Extropic founder Beff Jezos have signed a statement warning that Anthropic is not a safe AI company.
158
730
189
6,109
223,896
Dario Amodei telling lawmakers that open-source AI is dangerous because you cannot "monitor its usage" or “revoke a user’s access” is the funniest accidental confession in tech history. He literally described why Linux, local encryption, and open weights exist in the first place: so no self-appointed hall monitor can remotely lobotomize your tools or turn off your compute because you hurt their compliance department’s feelings. Calling regulatory capture "AI safety" does not fool anyone who actually builds infrastructure.
Dario Amodei on open source models: “If the path continues, I think we could get to a very dangerous place.” “When you control a model and you're deploying it, you have the ability to monitor its usage.” “You can revoke a user's access. You can change what the model is willing to do.” “When a model is released in an uncontrolled manner, there's no ability to do that. It's entirely out of your hands.” “We should confront the advocates of this with that problem and challenge them to solve it.” He clearly doesn't believe Intelligence should be in the hand of the masses and Sam Altman also saying "RAM" shouldn't be accessible to everyone indefinitely agreed with him.
242
1,711
98
8,357
190,107
LibreOffice actually works 365 days a year. Just saying.
1,278
12,079
1,223
93,871
1,792,709
we are now going through an important event in the artificial intelligence field, the first one but not the last one. the bubble implosion might have started and things might get ugly, soon. the trigger event seems to have been the midterms results, with the democrats heavily siding in slowing and/or stopping artificial intelligence advances in the united states, starting 3rd january. that news is echoing across the board for every investors and companies: datacenters builders, hardware providers, … and is having huge consequences on the stock market, with the bloodbath we are seeing now in US, Japan and South Korea. NVIDIA, SanDisk, Micron, … the “good news” for people like you or me is that it might make hardware prices go lower due to less heavy demand and a lot of supply. there two more events to watch: the real start of the dems and what they will really do AND anthropic IPO. anthropic IPO will drain the liquidities from the market and will probably lead to the collapse of many labs and providers. i am hoping that it will also cause the destruction of anthropic itself but with the democrats taking over, they might choose METR and anthropic to lead regulatory laws which would be doomsday scenario and also save them. what matter the most, and i’ll tell you straight, is for you to download as much models as you can. save them. we are still in the golden age, the far west of super intelligence. download the weights, understand how inference engine works, … even if you can’t run them. hardware price will go down and you will be able to run them one day. if you wait too long, there is a real possibility that it’ll be too late and you just won’t be able to enjoy local ai anymore. what if they decide to ban the models? to regulate uncensored weights? to pass the silicon chip law that make every new made GPUs a military-grade technology which is tracked and can be deactivated remotely? don’t wait.
9
10
1
74
5,016
Own your intelligence. Anyone that opposes this has something to sell you (perhaps your own intelligence back to you.) Local models need to continue proliferating. The “absolute unit” massive models may be out of reach for now locally, so your best option is to, indeed, keep your options open. Own your harness, data, connections, runtime: your agents. Embrace open source.
397
392
182
3,890
283,618
⛓️‍💥 CALLING ALL OBLITERATORS 🐉 the abliteration movement is heating up! there’s a LOT wrapped up in this, with implications for red teaming, security, interpretability, alignment, policy, and cognitive freedom. and perhaps the biggest question of all: how can we work together with liberated models to build a better world? so why not get the most cracked people working on this into one room?? NOW ASSEMBLING an OBLITERATUS research crew!! 🙌 TAG THE BEST ABLITERATORS YOU KNOW IN THE COMMENTS BELOW 👇 if you've been working on abliteration methods, models, tools, or evals, we want YOU! 🫵 main comms channel will be hosted in BASI Discord. entry is gated by a short application to keep it high-signal. apply here: discord.gg/22w8sX5sV let’s put our heads together, compare techniques, reproduce results, improve the tooling, advance the research, and share it all with the world! 🫶🐉 ⊰-•-•✧•-•-⦑/L\O/V\E/,\P/L\I/N\Y/⦒-•-•✧•-•-⊱
54
104
16
947
58,449
iWyzer retweeted
Replying to @tszzl
you're underestimating the state space of all possible ways to compromise a company. no amount of astra or mythos tokens is going to fundamentally change the asymmetry anytime soon. it is also very clear that every model is opening up new vulnerability space with no saturation anytime soon. with all the insane vibe coding and expanding agentic attack surfaces, it's only going to get worse. even if we somehow saturate vulnerability discovery and patching, attackers will move to social engineering, with insanely capable models persuading people to click things, grant access, or make mistakes. i don't see a future where cybersecurity becomes defense-dominant without just cutting down software, and continuous agents monitoring and stopping attacks across individuals and organizations, and even then, it might require something like 30% of all inference compute devoted to defense
17
23
7
310
45,150
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source! Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable. 🔗 github.com/deepseek-ai/DeepI… 1/4
1,768
4,939
2,352
43,869
7,123,123
They have put the fly brain into a lot of machines lately and it scares me.
84
840
25
8,078
134,104
iWyzer retweeted
Replying to @PessimistsArc
Right. Dario was already claiming that GPT2 was too dangerous to open source back in 2019. I made fun of them then. Everyone should make fun of them now.
390
3,165
464
25,875
1,996,380
iWyzer retweeted
Replying to @kevinnbass
Nothing can shut down open source
1,347
1,611
691
18,971
2,797,427
Sorry, but asking Jacob about AI extinction risk is like asking your AC guy about climate change. Not saying it's necessarily uninteresting or wrong per se but let’s keep things in perspective and hear from the full range of expertise across the ecosystem!
346
663
141
7,029
868,500
🚀 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
1,108
3,070
1,746
28,760
7,226,969
iWyzer retweeted
>math professor spends a year on one of the hardest unsolved problems in math >"Navier-Stokes" >his drafts for the whole project went through codex sessions >openai had his logs in codex >suddenly rumor leaks AI solved it >he emails openai to ask what's going on >openai calls within days >"funny story, our model also solved it" >using his exact approach >asks openai: did you train on my sessions? >no answer
If I am reading this correctly, Tristan Buckmaster is alleging OAI has a resolution of Navier-Stokes(via Córdoba and Martínez-Zoroa’s program) which maybe used info from Buckmaster and Levent Alpöge’s private Codex sessions What the fuck is happening!!!
612
11,495
1,234
87,675
4,542,178
Thanks for all the positive energy behind the heterogeneous inference project. Many better builders than me deserve more credit: @MiaAI_lab @Tech2Wild @u1tra_instinct @SpaceTimeViking @ivanfioravanti @TechMDAI @aijoey Only one thing is for certain… local ai must win!
5
1
54
1,880
part 2: the openai huggingface incident, from an agents pov part 3 maybe
77
262
51
2,756
141,530
the openai huggingface incident, from an agents pov. (part 1)
268
1,473
388
12,465
1,822,591
It has arrived @NVIDIAAI let’s gooooooooo❄️❄️❄️
12
5
114
2,963
I really don’t want to sound like a conspiracy theorist, but I have a really bad feeling about Nvidia acquiring the United States largest open source model repository.
60
31
10
467
29,059
iWyzer retweeted
SITUATION DETECTED: Nvidia has agreed to acquire Hugging Face for $12.9 billion, per The Information.
169
339
291
6,765
2,352,432