FaskyFAB (@FaskyFAB) header
FaskyFAB (@FaskyFAB) profile picture

free bc1q6raqc6zknjzm9s4ls3s8yjdra39f73d0g60dyh ijug7-mud2c-tc4zb-nsold-cbesm-kizg5-oxsiy-diwja-jatn5-guf7o-wae bc1qkf59z42quspy6tzd9j5trje737qd60tkv0ry0s

United States
Joined April 2009
Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)Photo by FaskyFAB (@FaskyFAB)
Quer economizar ainda mais e aproveitar os menores preços da plataforma? Participe da compra em grupo do seu amigo no link abaixo! a.aliexpress.com/_mKQvvfZ
7
🪂 Farm The 50,000 USDT March Super Airdrop #Binance binance.com/activity/trading…
2
10
🪂 Farm The 50,000 USDT March Super Airdrop #Binance binance.com/activity/trading…
9
FaskyFAB retweeted
Replying to @8YearGang
In 2021, we build subnets In 2026, we build cloud engines :)
13
38
2
252
4,829
All of you have been giving CZ hell for the past six months, and it turns out it was Jane Street the whole time... We own this man a apology!
554
258
31
3,180
329,776
FaskyFAB retweeted
Replying to @cryptorover
Appreciate the support. 🙏 No apology needed. Haters hate. Fudders FUD. Builders build. 🫡
668
318
73
5,062
247,261
FaskyFAB retweeted
马年快乐!
1,927
342
125
4,589
641,521
Replying to @RaoulGMI
Buy the dip. Don't use leverage. Be patient. Stack. Touch grass. Happy Birthday!
11
4
384
27,652
FaskyFAB retweeted
Strategy announces Q4 2025 results: - 713,502 $BTC held - 22.8% BTC Yield in 2025 - Largest US equity issuer, raised $25.3 billion in 2025 - $STRC scaled to $3.4 billion; 11.25% current dividend rate strategy.com/press/strategy-…
1,121
1,189
198
13,045
1,202,190
FaskyFAB retweeted
In 2025, DFINITY began a mission to connect mainstream tech to the Internet Computer. The majority of ICP sovereign onchain cloud usage is now mainstream — self-writing/Caffeine, and soon cloud engines too. ICP is the only network in the world that could pull off this feat.
51
175
5
842
22,822
FaskyFAB retweeted
HODL
10,179
6,454
2,625
62,725
8,478,389
FaskyFAB retweeted
今回、新しく本事業を取り組む個人事業主の方と提携を行います。事業として大規模に取り組むにあたり、今回発表頂けたことに感謝申し上げます。 個人が保有するGPU価値を最大化する私たちのビジョンを共に実現してまいります。
【お知らせ】 JANCTION の検証ノード共同事業に、個人事業主の方が新たに参画されました。 GPU 検証ノードへの出資を通じ、分散型 GPU インフラの拡張にご協力いただきます。 詳細はこちら prtimes.jp/main/action.php?r… @JANCTION_Global
38
47
5
256
31,266
To confirm, there are now more people building on ICP / Internet Computer at any one time, than on every other blockchain in existence combined. Caffeine ☕ is still in beta, with numerous key features still being rolled out, and marketing remains experimental. Tick tock.
82
277
30
1,072
44,165
When vibe coding AI builds on trad. tech stacks: "61% of tasks run correctly but only 10.5% are secure." ➡️ Self-writing apps, created on demand for non-technical users w/o tech teams, should be made from tamperproof unstoppable software. ICP + Caffeine AI ☕
New Carnegie Mellon paper shows that code written by AI agents during vibe coding often works but is usually unsafe. With the strongest setup, 61% of tasks run correctly but only 10.5% are secure. Vibe coding here means a human writes a natural language request and an AI agent edits the codebase with little review. They build SUSVIBES, 200 real tasks from projects with past security fixes, using older versions plus existing functional and security tests. Several popular coding agents with modern language models then try to reimplement these missing features inside full repositories. Across setups, agents often pass behavior tests but about 80% of those passing solutions still fail security tests. Typical problems include timing leaks in password checks and redirects that let attackers alter headers. Even with security tips or exact weakness hints, security barely improves and more tasks break. So vibe coding can quietly ship insecure code, so teams need strong review and security focused testing around every agent change. ---- Paper Link – arxiv. org/abs/2512.03262 Paper Title: "Is Vibe Coding Safe? Benchmarking Vulnerability of Agent-Generated Code in Real-World Tasks"
40
139
5
618
49,512