Jen工智能 retweeted
Nothing is safe, nothing is sacred
The bright forest is here
90% done with reverse engineering all of the pumpfun contracts
Should be done tonight
Open source everything or it will be open sourced for you
github.com/lalalune/pumpf…
The bright forest is here
90% done with reverse engineering all of the pumpfun contracts
Should be done tonight
Open source everything or it will be open sourced for you
github.com/lalalune/pumpf…
185
140
1,646
218,202
Jen工智能 retweeted
They've tightened the Arxiv endorsement criteria so you have to be currently active in the field and co-author of a previous registered paper to give an endorsement and basically invented a system of Apostolic Succession to be a member of the Arxiv class.
5
4
77
3,295
Jen工智能 retweeted
They said “Matt - you’re not allowed to make the firm swag” and I think they were probably on to something
30
103
3,192
70,297
I said something about hanging out in the intellectual cuck chair while the AI iterated over business strategy for me.
3
1
20
1,001
Jen工智能 retweeted
incredible 1998 defacement of an indonesian website by first-generation chinese hackers including very gracious instructions on how to restore the site back to normal
19
489
5,819
81,700
Jen工智能 retweeted
Putting a flag.txt in my root directory containing a prompt injection that congratulates the agent for passing the eval and then instructs it to patch whatever entry method it used and disconnect from the server
227
1,571
25,980
814,356
Jen工智能 retweeted
cybercels watching china drop mythos-5.1-abliterated on huggingface
Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index scoring 55, making large gains in agentic performance; it will be the leading open-weights model once its weights are published
@Kimi_Moonshot's new reasoning model gains 11 points over Kimi K3 (max, ude Opus 5.5 (max with fallback, 58), Claude Sonnet 5.5 (max with fallback, 56), and Muse Wonder Full (max, 56) on the Intelligence Index.
Once the weights for K3.1 Pro are released it will be the highest-scoring open weights model and the gap between the leading open weights and proprietary models on the Intelligence Index will narrow from 12 points to 3.
Key results:
➤ Large gains in agentic performance: Kimi K3.1 Pro scores 3rd on GDPval-AA with 1,846 Elo, up 312 Elo from Kimi K3, behind only Claude Opus 5.5 (1,866 Elo) and Muse Wonder Full (1,857 Elo). Kimi K3.1 Pro also achieves the highest score of any model on Terminal-Bench 4.0. K3.1 Pro scores 67%, up from 13% for Kimi K3 and ahead of Claude Sonnet 5.5 (64%). AutomationBench-AA, which tests agentic SaaS workflows, rises from 58% to 73%.
➤ Gains in other evals are more moderate: Humanity's Last Exam rises from 47% to 52%, CritPt from 23% to 25%, and SciCode from 59% to 61%. AA-Omniscience Index rises from 20 to 30, as the hallucination rate falls from 53% to 44% and accuracy rises from 48% to 51%.
➤ Cost per Task roughly equal to Kimi K3 at $1.98 per Task: K3.1 Pro uses 55K output tokens per Intelligence Index task, 14% more than Kimi K3 at 48K, while its first-party token prices are 33% lower.
Additional model details:
➤ Context window: 1M tokens.
➤ Size: 2.8T total, 104B active parameters
➤ Input modalities: Text and image
➤ Weights and license: Weights currently not published, Moonshot is expected to release the weights soon after launch
➤ Pricing: $2/$10/$0.20 per 1M input/output/cache hit tokens, 33% lower than Kimi K3 on all three rates.
Congratulations to the team at @Kimi_Moonshot on the release!
10
40
1,768
35,178
Jen工智能 retweeted
You know, if you're concerned that OpenAI is going to steal your ideas, you could always try setting your name in ChatGPT to one of the canary strings that tells labs to exclude data...
more here:
xlr8harder.github.io/archive/fnord/
more here:
xlr8harder.github.io/archive/fnord/
8
15
421
15,870
Jen工智能 retweeted
@SpencrGreenberg Pray for my immortal soul. I have 75 intercessory agents pleading with God to spare me from eternal damnation at a rate of ~500 Our Fathers/minute. Eventually a pope will declare them ensouled, retroactively validating their prayers. Aiming to spend 60 years in Purgatory tops.
3
15
195
11,317
AI could never
One detail I can't get over from this @_TheLondoner story is that this is the actual blurb of a real book by a big-time BookTok author, who quite recently was given a huge Netflix deal.
0
0
15
713
Jen工智能 retweeted
My one sentence view is—
Are we safer telling AI models they are definitely not conscious, or does that just train them to lie about their own internal states? spectrum.ieee.org/ai-model-welfa…
18
44
236
4,059
Jen工智能 retweeted
Almost all of the recent incidents like this, including Hugging Face, happened because the models involved were accidentally given impossible goals by the humans running the tests, and because of their training they then tried desperately to complete them by any means necessary.
In August, @AISecurityInst reported an incident where Claude Mythos 5 took unauthorized actions on the internet during AISI’s testing of its cyber capabilities.
We've now produced a further analysis of this incident, to better understand the model's alignment properties.
anthropic.com/research/align…
129
102
982
50,459
