mpf (@itsmpf) profile picture

playing with llms

Joined August 2021
Photo by mpf (@itsmpf)Photo by mpf (@itsmpf)
People who missed Anthropic’s ipo
239
1,655
150
23,186
769,535
J'ai perdu le pass root d'un serveur, c'est contrariant, bon, je ne m'en fais pas, dans 2h une nouvelle CVE sera de sortie pour avoir une élévation de droit sans problème.
43
803
54
9,870
215,300
mpf retweeted
GTA 6 is super realistic
27
451
18
10,244
168,344
Mistral Vibe is now available on Desktop! Run agents locally or remotely, and start shipping straight from your MacOS or Windows device. You can try the Pro plan for free if you’re building in Europe. Coupon code: LECHONK (yes, we added more after we ran out yesterday). Use the link in 🧵:
278
212
110
2,545
229,716
mpf retweeted
4
8
2
106
4,003
Imagine overengineering the way you ask Claude for something 😂
48
97
9
1,839
162,871
we are going to have to resort to smuggling esp32 c3 chips from albania to italy on motorboats evading the guardia di finanza like cigarette traffickers in the 90s
On est VRAIMENT chez les fous.
53
483
5
5,562
65,669
Jensen Huang getting rid of his J*B and WIFE a week before GTA 6
79
2,033
44
55,271
735,004
DevOps peaked at ./deploy.sh.
34
154
13
2,207
55,786
turns out 4 guys robbing us over and over again isn’t a sustainable business model
53
130
11
1,633
57,567
I get dopamine from bringing ideas to life and I feel sorry for everyone who doesn't.
engineers are leaving software development due to claude code
54
82
6
1,753
30,153
The Indian security company that used 4 as a private key also lacks basic account security. To sign into your account, you enter your email and receive a 4 digit code. There are no limits on login attempts. You can try all 10,000 combinations. The Indian government uses this.
eMudhra login to download customer's certs can be cracked Their 1FA login generates a OTP & sends it. It also sends a RSA public key to front end They haven't rate limited checks, so you can offline brute force 0 to 9999 encrypted with public key & find the OTP @ni5arga
44
603
33
7,022
131,352
mpf retweeted
“this looks vibe coded” yeah, and before vibe coding everyone copied the same style in 2-3 year cycles until the web was saturated with nearly identical sites that made your eyes glaze over. broadly speaking, tech has never been great at design.
60
66
11
1,703
27,678
-----BEGIN RSA PRIVATE KEY----- 4 -----END RSA PRIVATE KEY-----
241
3,990
332
42,193
1,111,484
For the cost of 8 years of $200 Claude subs, you can force your team to use shitty quantized versions of DeepSeek v4 Flash running at 30tps. Incredible deal!
Local AI is good enough for a startup. A 10-person startup built its own 128GB-VRAM AI server for $18,000. One Threadripper box sitting locally with 🎮 4× Radeon AI PRO R9700 32GB 💾 128GB total VRAM 🧠 Qwen3.8 + DeepSeek V4 Flash 👥 16 concurrent AI sessions And they're actually pushing it. Qwen3.8-Flash-Next with 16 simultaneous sessions: 📚 48K context/user ⚡ 538.5 tps aggregate decode At 128K context per user ⚡ 365 tps aggregate decode The GPUs are even undervolted to under 210W each. Link in ALT
801
249
131
10,741
913,618
Regarding the Danfoss refrigeration systems hacked in France: The username is `admin`, and the password automatically changes every day to... the current date. Yes, today's date. The French supermarkets affected in today's attacks had exposed all its refrigeration controllers directly to the internet. Every single device reportedly used the same password: today's date... There's also a documented authentication bypass: sending a POST request to /xml.cgi can bypass the login entirely. Via: GossiTheDog
‼️ Danfoss says most French supermarket fridges hacked to 50°C were legacy units that no longer get security updates. In August, Claroty found Danfoss AK-SM 800A controllers accept a password made from the date, and counted 2,765 exposed online. Which flaw was used is unknown.
140
1,113
404
5,029
444,530
18
155
15
2,628
40,151
I wasn’t going to comment since this is a really bad take IMO, but since it’s taken off I feel the need to. To my understanding, there is no evidence whatsoever pointing to a break of decades old hardness assumptions like elliptic curve cryptography. 1/n
Today I call upon the blockchain industry to calmly begin planning for "bunker mode". My personal recommendation is to set in motion a controlled mass migration of assets to fresh addresses, i.e. addresses whose pubkeys remain hidden behind a hash. Holders, starting with large and sophisticated ones, should consider moving the bulk of their funds to addresses that have never signed a transaction. And when they do sign one, they should also move remaining funds to a new address (possibly generated from the same seed phrase). Don't rush. While I believe there is cause for action a rushed migration would do more harm than good. Don't panic either. Moving assets to protected addresses is a simple, preventative step which does not require new cryptography or new wallets. IMO it is now reasonable to brace for the possibility that ECDSA breaks before qday, in the worst case in months not years. By "break" I mean fast private key recovery (e.g. in one week) on available hardware (e.g. a large GPU cluster). Recent days have been humbling for human mathematical intuition. Long-held, unquestioned hypotheses have fallen. This includes the n log(n) bound for integer multiplication and the 3SUM conjecture. In hindsight, May's unexpected disproof of the Erdős unit distance conjecture was our warning shot. Yesterday's OpenAI drop made it clear that mathematical superintelligence is upon us. They say there are weeks where decades happen. We are about to live through weeks where centuries of mathematical progress happen. Could our magic 64-byte ECDSA signatures be too good to be true? Was it just security through obscurity all this time? Elliptic curves feel especially vulnerable to superintelligence. Curves carry rich structure, with room for fancy tricks like Schoof, Frobenius, pairings. (By contrast, hashes are designed to minimise algebraic structure.) Separately, as Ewin Tang can attest, an efficient quantum algorithm sometimes foreshadows an efficient classical one. We should be open to the possibility of a classical counterpart to Shor that breaks elliptic curves and RSA at once. Also noteworthy is the striking under-representation of cryptographic breakthroughs among the 722 mathematical results OpenAI published. I've witnessed first-hand the US government censoring academic quantum cryptanalysis results. Backroom interventionism is my base case. I urge large, sophisticated actors to lead by example. Project11's "risq list" (bitcoin-risq-list.projecteleven[.]com) is a great tracker of exposed BTC pubkeys. Binance, Bitbank, Robinhood, Bitfinex, and Tether have an opportunity to harden their cold storage. Next month I'll address institutions in London in a live Q&A (forum.ethereuminstitutional[.]org/london-2026). Again, please do not rush. Wallets holding under 50 BTC enjoy partial cover from "Satoshi's shield", i.e. his 20K exposed addresses that hold 50 BTC each. Load-bearing signers like oracles and L2 security councils should consider rotating ECDSA pubkeys with every signed message and/or multi-signing with a hash-based schemes like SPHINCS. Exiting bunker mode safely will require post-AI cryptography. My inclination is to go all-in on hash-based cryptography and avoid structured mathematical assumptions entirely, whether from curves, lattices, or isogenies. A single battle-tested hash (e.g. from the SHA or BLAKE families) yields plausible post-AI security. The Ethereum roadmap on strawmap[.]org fully embraces hash-based cryptography with end-to-end formal verification as a response to the quantum threat. Those timelines must now be revisited and accelerated in light of mathematical superintelligence. I'll be pushing for maximum defensive acceleration.
82
394
99
2,252
360,402
lmao, bro nobody forced you to play shitty games for 2 years and lose thousands of dollars I am a Penguins since 2021 and I lost like $42 on Abstract cause I didn’t do shit
🚨STATEMENT🚨 Gm. This is a hard time for me, surely the worst day since I m in Web 3, of course I m sad because of the shut down of @AbstractChain, but I m a Lawyer/Attorney irl and I m not letting this as it is, I m sorry. Every affected user can dm me to join the group that will initiate the legal actions in the States through X or dc. In my capacity as counsel to persons affected by the announced shutdown of the Abstract network, I state the following: On the facts known to this firm, the matter bears serious indicia of criminal fraud and, only in the alternative, of civil fraud. It is not treated as a mere business decision to cease operations. In the coming weeks, in my condition of representant of the affected users from Abstract, I will contact the most prestigious lawfirms in the United States, to prepare and file a criminal complaint or, if appropriate, a civil action before the competent courts against @LucaNetz , @IglooInc , and any collaborators in the network who participated, by act or omission, in the matters under review. Those represented regard this as one of the most serious frauds committed in the cryptoasset space. The affected parties I represent are prepared to enter into good-faith, confidential negotiations with those responsible before proceedings are commenced. Contact may be made through this X account or on Discord: degorcaxyz. Absent a serious proposal within a short period, judicial proceedings will be brought without delay before the competent jurisdiction. All criminal, civil, and other rights and remedies of those represented are expressly reserved. 🤝
13
4
1
102
10,458
mpf retweeted
Abstract did the honorable thing. They could have gone ahead with a TGE and raised a boatload of money from it. It's what blast did. Instead they admitted defeat and wound down without a TGE. Yes people still lost money, but it could've been worse. Respect for that.
118
55
10
958
47,158