Web3 GTM 0 to 1 Marketing Strategist since '20 Scaled Web3 App to 300K+ Users & 5K+ DAU --- Editor-in-Chief @taodaily_io ---
Church of RAO
Joined September 2013
- Tweets9.9K
- Following475
- Followers1.3K
- Likes16K
Pinned Tweet
We've stepped up our editorial operations at @taodaily_io, and one area we're putting more focus on is analytics.
Expect more data-driven coverage to help the Bittensor community better understand what's happening across the ecosystem.
Big thanks to @subnetradarcom for making valuable data accessible and supporting our work.
More to come.
Top 10 Alpha Tokens Held by Bittensor’s Top 100 Whale Wallets
Lium (SN51) | @lium_io — 211,928 τ
Chutes (SN64) | @chutes_ai — 98,957 τ
Engy (SN53) | @engyai — 96,931 τ
Targon (SN4) | @TargonCompute — 91,377 τ
SayGM (SN28) | @say_gm_ — 75,322 τ
Affine (SN120) | @affine_io — 45,130 τ
Actual (SN95) | @actualinc — 43,376 τ
Vanta (SN8) | @VantaTrading — 34,218 τ
Score (SN44) | @webuildscore — 30,990 τ
Hippius (SN75) | @Hippius_cloud — 28,488 τ
Source: @subnetradarcom
σℓαigє τ retweeted
Finding new research papers is easy. Knowing which ones people are paying attention to is harder.
Desearch’s Arxiv Chatter Digest searches recent papers, pulls their abstracts, and tracks related discussions on X.
In one example, the full process cost about $0.06 👇
taodaily.io/desearchs-arxiv-…
σℓαigє τ retweeted
We see a growing trend of subnet intelligence as an enabling layer inside products, rather than buyers consuming the raw intelligence commodity directly.
Yanez's new Gate and Pulse tools are an example: proof-of-humanhood intelligence from Bittensor, productized by the subnet operator into a polished toolkit for sensitive agentic financial transactions. $TAO
Building an agent that can spend money? Check what it has access to.
Earlier this month, a poisoned npm package used in AI agent tooling went after .env files, wallets and AI tool configs. If your agent holds your payment information, whatever compromises it can also spend it.
With Yanez Gate and Pulse, your agents can request a transaction but never touch the funds. Your phone buzzes, you approve with your face. Your face scan never leaves your phone.
Gate keeps your money behind a smart contract your agent can't open on its own. Pulse is what unlocks it.
No fees through Nov 30.
Try it out: docs.yanez.ai/pulse/showcase…
Interesting about what @carbonphysicsai is doing and their appearance in Revenue Search 👏
🏗️ Carbon targets engineering customers for its physics AI
Carbon (SN116) plans to turn its first physics-AI competitions into models that engineering firms can test and potentially buy. Ryan Bequette says the team is starting with realistic customer problems—not just contests to see which miner can produce the highest-scoring model.
🔑 Key points
🔹 Four or five challenges planned: Carbon intends to use early engineering challenges to test model accuracy, identify failure points, and assess whether the results are reliable enough for engineers.
🔹 Smaller firms come first: The initial target is engineering teams of roughly 10–50 people, including specialists working on problems such as cooling systems—not major aerospace companies or government agencies.
🔹 Models accelerate simulations: Carbon’s early product would be a specialized physics model that speeds up or replaces part of an existing simulation process, helping engineers test more design options.
🔹 Customer IP stays protected: Bequette says paying customers would control and use models developed for them, while miners may receive performance feedback without necessarily receiving the winning model’s weights.
🔹 Pricing is still unproven: Carbon is using its first challenges to estimate the cost of producing useful models. Bequette mentioned comparable bespoke work in the mid-five to low-six figures, but said early pricing will likely depend on each customer.
🔹 Broader products may follow: If the first models prove useful, Carbon could explore reusable engineering products, design software, and more complete AI-assisted engineering services.
🔎 Why it matters
🔹 Carbon is testing whether subnet competitions can produce specialized tools that solve real commercial problems—not just strong benchmark results.
🔹 Starting with smaller engineering teams could make early adoption more achievable, but customer interest still needs to turn into paid projects.
🔹 The key proof points are model accuracy, engineering trust, repeatable delivery, and sustainable pricing.
🎯 Bottom line: Carbon’s next phase is about validating whether its physics-AI models can fit into real engineering workflows and win paying customers. The opportunity is significant, but commercial traction depends on results from the first challenges—not just the subnet’s ambition.
#Bittensor #Carbon #SN116 #PhysicsAI #Engineering #DecentralizedAI
taodaily.io/ryan-bequette-ex…
σℓαigє τ retweeted
You’ve probably seen a lot of updates about Grayscale’s Bittensor SEC filings lately.
If you’ve been finding it hard to keep up, we put everything together in one simple breakdown covering what has happened, where the proposed Bittensor ETF currently stands, and what to expect next.
taodaily.io/grayscale-submit…
σℓαigє τ retweeted
BITTENSOR THIS WEEK: Key Updates You Need to Know
This week features updates from subnet teams like @lium_io (SN51), @affine_io (SN120), @say_gm_ (SN28), @cacheon_ai (SN14), Ingot (SN97), @BitMindAI (SN34), Teutonic (SN3), @reaxlabs (SN92), and @trishoolai (SN23).
#Bittensor $TAO
σℓαigє τ retweeted
Here are the Bittensor subnets where a single large outside holder could have the biggest impact on price if they exited at once.
The estimate excludes subnet owners, protocol accounts, locked alpha, and miner collateral.
Source: SubnetRadar
σℓαigє τ retweeted
.@CrucibleLabs Wallet just made its mobile apps a lot more capable.
The full Chrome feature set is now available on iOS and Android, including reward allocation, subnet discovery, token data, and Smart Accounts.
It also expanded iOS access to Argentina, Japan, and Brazil 👇
taodaily.io/crucible-wallet-…
σℓαigє τ retweeted
Are you developing agents for a bank? For a corporation? Save this.
US bank regulators are asking banks five questions about the agents they use (Reuters, Oct 13). A bank customer will pass them to you, and other enterprises are starting to ask the same:
1. How are the agent's actions limited, monitored and approved?
2. Can each decision be tracked and audited?
3. What keeps it within its authority?
4. Where do humans stay in key decisions?
5. How do you shut it down?
1, 3 and 4 share one answer: the agent holds no credential, and a named person approves the sensitive actions, recorded outside the agent.
σℓαigє τ retweeted
I made this video to explain what @perturbaix (Perturb SN26) is building on Bittensor.
Perturb is one of the gems in the ecosystem rn, and this video breaks down what they're building in a way anyone can understand.
I think more people needs to watch it, let's share it widely!
Everyone wants AI to get smarter. But who's making sure it gets safer?
As AI becomes more powerful, its vulnerabilities become more consequential.
Perturb (SN26) is exposing those weaknesses and helping build resilient AI.
@perturbaix #PerturbVideoCompetition #PerturbVideo
σℓαigє τ retweeted
Bittensor 24H Market Snapshot
🔹 BitMind (SN34): Deepfake and synthetic media detection.
🔹 Quantum Compute (SN48): Marketplace for real quantum circuit execution.
🔹 TrajectoryRL (SN11): Competition for agent scaffolds on small open LLMs.
🔹 Almanac (SN41): Prediction market research and forecasting intelligence.
🔹 Lium (SN51): Decentralized GPU cloud for renting NVIDIA GPUs.
🔹 SayGM (SN28): Private gateway to frontier LLMs via TEEs.
🔹 Bitcast (SN93): Decentralized creator marketing and brand content marketplace.
🔹 Engy (SN53): Verified open-model inference on consumer GPUs.
🔹 THOR (SN112): Open robotics data marketplace for physical AI training.
🔹 ReadyAI (SN33): Turns raw data into structured AI-ready datasets.
🔹 Synth (SN50): Synthetic price path forecasting for financial markets.
🔹 Nodexo (SN106): Verified GPU rental marketplace with conviction staking.
🔹 Sundae Bar (SN121): Enterprise AI platform for skills and agent solutions.
Data source: @subnetradarcom
I made this video to explain what @perturbaix (Perturb SN26) is building on Bittensor.
Perturb is one of the gems in the ecosystem rn, and this video breaks down what they're building in a way anyone can understand.
I think more people needs to watch it, let's share it widely!
Everyone wants AI to get smarter. But who's making sure it gets safer?
As AI becomes more powerful, its vulnerabilities become more consequential.
Perturb (SN26) is exposing those weaknesses and helping build resilient AI.
@perturbaix #PerturbVideoCompetition #PerturbVideo
σℓαigє τ retweeted
Another day, another hack. A live example of what Part 1 is about.
Railway, a platform developers use to host apps, disclosed a breach on Oct 21. According to its preliminary report, the attacker got into its build system, took credentials for its CI environment and decrypted part of the settings customers store there (environment variables, where API keys and passwords usually live). Railway says at most 1.85% of workspaces were affected. It can’t tell which keys were decrypted, so it is asking everyone who signed up before Oct 21 to rotate all of them. Railway says the attacker was after AI provider keys.
Some customers also kept wallet private keys in those settings. Taylor Monahan (@tayvano_) traced a sweeper that started draining wallets at 19:36 UTC on Oct 21: 500+ transactions to 200+ addresses on 19+ chains. She ties the drains to keys stored on Railway. Railway hasn’t confirmed that part.
The owners didn’t leak their keys themselves. The keys were safe only as long as the platform storing them, its build pipeline and everything else with access stayed clean. Once the attacker had the keys, the blockchain processed their transactions exactly as it would the owners’, because a valid signature is all it checks.
If you hold funds with a key stored on a server, rotate it now and move the funds. Longer term, a key that can move money shouldn’t sit anywhere a breach like this can reach, and moving funds should need the owner’s own approval on a device they control.
σℓαigє τ retweeted
Everyone wants AI to get smarter. But who's making sure it gets safer?
As AI becomes more powerful, its vulnerabilities become more consequential.
Perturb (SN26) is exposing those weaknesses and helping build resilient AI.
@perturbaix #PerturbVideoCompetition #PerturbVideo
Interesting – at Ibadan, Nigeria
Subnets With the Deepest Pools
Chutes (SN64) | @chutes_ai — 200,765 τ
Lium (SN51) | @lium_io — 182,664 τ
Targon (SN4) | @TargonCompute — 136,229 τ
Vanta (SN8) | @VantaTrading — 84,817 τ
Affine (SN120) | @affine_io — 73,602 τ
Teutonic (SN3) — 71,992 τ
Score (SN44) | @webuildscore — 65,662 τ
IOTA (SN9) | @IOTA_SN9 — 53,178 τ
Gradients (SN56) | @gradients_ai — 50,154 τ
NOVA (SN68) | @metanova_labs — 39,640 τ
Source: SubnetRadar
SN59.
Subnets With the Fastest Holder Growth in the Last 7 Days
Lium (SN51) | @lium_io — +140 holders
Vanta (SN8) | @VantaTrading — +138
THOR (SN112) | @THORMachines — +131
REAX (SN92) | @reaxlabs — +123
SayGM (SN28) | @say_gm_ — +110
Subnet 59 (SN59) — +97
RedTeam (SN61) | @_redteam_ — +82
IOTA (SN9) | @IOTA_SN9 — +80
DaniLabs (SN2) | @danilabsai — +56
Targon (SN4) | @TargonCompute — +50
Source: SubnetRadar
σℓαigє τ retweeted
Insurance claims run on photos. So what happens when the photo is lying?
I built my #PerturbVideo submission around that question, using Angle 3: The claim that never happened
A real example was reported by The Guardian in May 2024. A van photo was taken from a tradesman's social media, edited to show a cracked bumper, and sent to UK insurer LV= with a fake invoice for more than £1,000. LV= had concerns and passed it to its fraud team. Allianz UK said claims involving app-edited images, videos and documents rose 300% between 2021-22 and 2022-23
My video then plays out a clearly labeled, made-up scenario where an automated claims model approves a doctored photo. A few tuned pixels that a person would never notice can leave a model wrong and still confident
The cost doesn't just vanish. Insurers cover the payout, and honest policyholders feel it in their premiums. The Coalition Against Insurance Fraud estimates US insurance fraud at $308.6 billion a year in its 2022 study, which is an industry figure and not something Perturb measured
Perturb @perturbaix flips the usual order and breaks the model first, before a fraudster gets the chance. On Bittensor Subnet 26, miners search for the smallest change that makes a model's answer flip. Validators confirm the flip is real and the change stays invisible, copied work gets no reward, and whoever finds the best attacks earns the most $TAO
For the team behind the model, that testing turns into a robustness score, a vulnerability heatmap, verified attack data to retrain on and an on-chain certificate of testing. Details live at perturbai.io
As of now, image-model testing is live and the network has been running since May 2026. Text, LLM and multimodal attacks are still on the roadmap, and November presales are a target rather than a done deal
σℓαigє τ retweeted
Bittensor Gamma tokens:
It's like 'Sign In With Google' for ALL subnet products.
No signup / onboarding. No KYC. No credit cards. No API keys.
Any agent with a $TAO wallet can ALREADY access all 128 subnets. The wallet IS the API key.
Standardized payment rails + access.
Interesting stat 👀
Subnets With the Fastest Holder Growth in the Last 7 Days
Lium (SN51) | @lium_io — +140 holders
Vanta (SN8) | @VantaTrading — +138
THOR (SN112) | @THORMachines — +131
REAX (SN92) | @reaxlabs — +123
SayGM (SN28) | @say_gm_ — +110
Subnet 59 (SN59) — +97
RedTeam (SN61) | @_redteam_ — +82
IOTA (SN9) | @IOTA_SN9 — +80
DaniLabs (SN2) | @danilabsai — +56
Targon (SN4) | @TargonCompute — +50
Source: SubnetRadar