Pinned Tweet
Can you prove you’re human? Join me LIVE with @josercaldera from @yanez__ai to explore deepfakes, human verification, and the latest updates from $TAO subnet SN54. Bring your questions 👇 x.com/i/broadcasts/1…
3
13
32
1,620
Shizzy retweeted
In KubeTEE, when a pod starts, it does an attestation that Confidential Computing is active and trusted by Intel Trust Authority. Only then does the pod receive the secrets from the KBS that gates it.
‼️ UPDATE: Hacker "k0t" breached 924 more Fortinet firewalls than first known, Hudson Rock says.
They hold 40,573 cleartext secrets, including 1,576 Active Directory bind passwords.
0
4
11
470
Shizzy retweeted
If you want to have a good overview over the LLM competition situation, read this to the end.
Interesting detail:
Open Weights models is falling behind for LLMs (see image), but the situation looks different Decision Models (aka Jev class models), which are way smaller and thus or small teams.
Interesting detail:
Open Weights models is falling behind for LLMs (see image), but the situation looks different Decision Models (aka Jev class models), which are way smaller and thus or small teams.
Announcing Artificial Analysis Intelligence Index v5, featuring 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets
The Artificial Analysis Intelligence Index is our composite measure of model capabilities across tasks sage today. v5 is our biggest upgrade since v4 launched in January. With v5, 60% of the Index is now based on evaluations with private test sets or held-out answers.
Alongside Intelligence Index v5, we’re launching the Artificial Analysis Progress Index v1, which tracks model capability over time on one continuous scale.
New to the Intelligence Index:
➤ FrontierCode 2.0: Cognition's newly-launched private benchmark of real-world coding tasks, graded on whether a repository maintainer would merge the agent's change
➤ Terminal-Bench-Science: 70 agentic scientific research tasks, run in a terminal environment
Upgraded:
➤ HLE-Diamond (replacing Humanity's Last Exam): a refined 1,000-question version of Humanity's Last Exam, split evenly between reasoning and knowledge questions
➤ AA-Omniscience v1.1: An updated dataset with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised
➤ CritPt v2: a private, 40-question hard subset of CritPt v2, revised by the CritPt authors and run with a code-execution tool
Staying in the Index: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0
Removed from the Index: GDPval-AA v2.1, SciCode, and AA-LCR v1.1
Key results:
➤ Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55): Anthropic and OpenAI hold the top five models. Gemini 4 Argon (high) is next at 50
➤ The open weights gap widens from 3 points to 12: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 46. Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32 - 26 points behind the leader and below Claude Haiku 5.5 (max, 36)
➤ Cost per task varies by more than 13x among the leading models: of the 10 highest-scoring models shown, GPT-6.1 Terra (max) is the cheapest at $0.96 per Intelligence Index task and Claude Fable 5.1 (max) the most expensive at $13.04
1
2
7
585
Beyond the candles. Into the conviction. ⚔️
TOMORROW I’m live with @TAOTemplar to dig into Bittensor’s evolution, expressive mechanisms, and what deserves your attention beyond price action.
What has changed—and what still needs to prove itself? Let’s get into it.
Thursday, Oct 29 · 1 PM ET / 10 AM PT
Set your reminder & bring your questions 👇
youtube.com/watch?v=RnnWez…
TOMORROW I’m live with @TAOTemplar to dig into Bittensor’s evolution, expressive mechanisms, and what deserves your attention beyond price action.
What has changed—and what still needs to prove itself? Let’s get into it.
Thursday, Oct 29 · 1 PM ET / 10 AM PT
Set your reminder & bring your questions 👇
youtube.com/watch?v=RnnWez…
4
10
27
1,246
Shizzy retweeted
One note about "the tasks stay private to avoid contamination":
If you call a model via an API, the model provider knows the task.
It's not private anymore then.
If you call a model via an API, the model provider knows the task.
It's not private anymore then.
FrontierCode 2.0 is the latest update of @cognition's private coding benchmark. Launched earlier today, it comprises 132 tasks in real open-source repositories, graded on whether a maintainer would merge the agent's change. An attempt that fails any blocking check scores 0, and to avoid contamination. We run every task three times in Stirrup, our agentic harness.
Claude Opus 5.5 (max) leads at 46%, followed by GPT-6 Astra (max), GPT-6.1 Terra (max) and Claude Sonnet 5.5 (max), all at 42%. DeepSeek V4.1 Flash 1017 (max) is the highest-scoring open weights model at 28%, 18 points behind the leader.
1
2
13
621
Finding leads is one thing. Turning them into action is the real test. TOMORROW I’m live with @LeadpoetAI: the latest from Bittensor SN71, how the product works, and what comes next. October 28 · 11 AM ET / 8 AM PT. Set your reminder & bring questions: youtube.com/watch?v=q2pGNL…
3
4
32
1,556
@LeadpoetAI SCHEDULE UPDATE: My livestream with Gavin from @LeadpoetAI has moved to Thursday, November 5 at 11 AM ET / 8 AM PT. The October 28 time in the original post is no longer current.
Same YouTube link, new date: youtube.com/watch?v=q2pGNL…
Same YouTube link, new date: youtube.com/watch?v=q2pGNL…
0
1
3
227
Shizzy retweeted
🇪🇺🇪🇺🇪🇺 Special EU post 🇪🇺🇪🇺🇪🇺
Dear fellow europeans, if you need
- state of the art Jev-class models,
- at fast speed and
- low prices
- and you need it EU COMPLIANT
(built for GDPR and EU AI Act)
Please go here:
Dear fellow europeans, if you need
- state of the art Jev-class models,
- at fast speed and
- low prices
- and you need it EU COMPLIANT
(built for GDPR and EU AI Act)
Please go here:
Need decision models without your data leaving the EU?
Our EU tier:
• EU requests always stay in the EU
• Priority on our EU capacity
• Never sent to the peer-to-peer network
• Prompts and answers not stored or used for training
• DPA included
decisionmodels.io/eu
1
4
19
761
Shizzy retweeted
@yanez__ai @josercaldera @ShizzyUnchained Best up-to-date presentation of what Yanez is up to. Undervalued is an understatement.
1
2
4
256
Shizzy retweeted
Our founder @josercaldera joins @ShizzyUnchained for an in-depth look at SN54.
Jose goes into detail not just on Proof of Humanhood and it's use cases, but also dives into the wider vision behind it:
the advantage Bittensor brings it, the rise and role of agents, thoughts on where the space is going, and how what they have built / are building, meets the coming demand.
$TAO
Jose goes into detail not just on Proof of Humanhood and it's use cases, but also dives into the wider vision behind it:
the advantage Bittensor brings it, the rise and role of agents, thoughts on where the space is going, and how what they have built / are building, meets the coming demand.
$TAO
Can you prove you’re human? Join me LIVE with @josercaldera from @yanez__ai to explore deepfakes, human verification, and the latest updates from $TAO subnet SN54. Bring your questions 👇 x.com/i/broadcasts/1…
2
6
19
648
Shizzy retweeted
Great conversation with @ShizzyUnchained yesterday. If you missed it: humanhood, presence and uniqueness, and how SN54 miners generate thousands of images a day that challenge and help train our detection models.
@yanez__ai $TAO
@yanez__ai $TAO
DID YOU MISS THIS? 🤖
How do you prove there’s a real human behind an AI agent?
@josercaldera joined me to break down @yanez__ai, Bittensor SN54, deepfake defense and why identity matters for agentic payments.
Watch the full replay 👇
youtube.com/watch?v=STPM5B…
1
5
16
452
Bittensor subnets are connecting.
SN64 Chutes is now available through SN28 SayGM—a documented new route for confidential open-weight models. Access is real; adoption and privacy guarantees still need scrutiny.
Today’s news also covers model-comparison tools, drone simulation, subnet renames and verified allocations/burns.
Read the full October 28 edition:
shizzyunchained.com/subnet-news/20…
SN64 Chutes is now available through SN28 SayGM—a documented new route for confidential open-weight models. Access is real; adoption and privacy guarantees still need scrutiny.
Today’s news also covers model-comparison tools, drone simulation, subnet renames and verified allocations/burns.
Read the full October 28 edition:
shizzyunchained.com/subnet-news/20…
Chutes is now live on @say_gm_
SayGM gives you one API key for Claude, GPT, Gemini and open-source models.
When you pick one of SayGM’s private models, it can now run on Chutes, inside secure hardware where no one can read your prompts. Not even us.
See the full article here: chutes.ai/news/chutes-mo…
5
11
85
1,756
DID YOU MISS THIS? 🤖
How do you prove there’s a real human behind an AI agent?
@josercaldera joined me to break down @yanez__ai, Bittensor SN54, deepfake defense and why identity matters for agentic payments.
Watch the full replay 👇
youtube.com/watch?v=STPM5B…
How do you prove there’s a real human behind an AI agent?
@josercaldera joined me to break down @yanez__ai, Bittensor SN54, deepfake defense and why identity matters for agentic payments.
Watch the full replay 👇
youtube.com/watch?v=STPM5B…
2
5
21
1,929
Revenue funding alpha buybacks is worth watching. SN4 Targon reports 8,450 TAO-equivalent in buybacks/burns. I independently verified 152,314.359 alpha burned on-chain. The burn is verified; revenue attribution and TAO valuations remain team-reported. Full breakdown: shizzyunchained.com/subnet-news/20…
We have just repurchased and burned 1,772 TAO (~$445K) of SN4 alpha using organic revenues generated via Targon.com
We have also burned an additional 6,678 TAO (~$1.7M) worth of SN4 alpha, which also represents revenue generated via Targon
0
7
51
1,413
Shizzy retweeted
BREAKING: 🇺🇸 US Treasury just bought back $6,000,000,000 of its own debt.
Bullish for markets.
Bullish for markets.
114
177
1,658
85,338
Shizzy retweeted
We have just repurchased and burned 1,772 TAO (~$445K) of SN4 alpha using organic revenues generated via Targon.com
We have also burned an additional 6,678 TAO (~$1.7M) worth of SN4 alpha, which also represents revenue generated via Targon
We have also burned an additional 6,678 TAO (~$1.7M) worth of SN4 alpha, which also represents revenue generated via Targon
39
100
482
58,370
I am long bittensor:native and love bittensor's focus on scaling AI
I am long $TAO and love bittensor's focus on scaling AI
0
15
84
1,578
Beyond the red candles: what’s actually being built on Bittensor?
Harnyx has added REAX decision models for SN67 research miners. Beam has published EVM delivery receipts. Taoswap announced an Ethereum bridge.
Today’s news separates working tools, team claims and roadmap plans—plus owner-key changes, eTAO and verified burns.
Read the full edition: shizzyunchained.com/subnet-news/20…
Harnyx has added REAX decision models for SN67 research miners. Beam has published EVM delivery receipts. Taoswap announced an Ethereum bridge.
Today’s news separates working tools, team claims and roadmap plans—plus owner-key changes, eTAO and verified burns.
Read the full edition: shizzyunchained.com/subnet-news/20…
0
8
28
897