Irresponsibly long τao. Life. Liberτy. Biττensor.
Church of Rao, τaoτardisτan
Joined September 2021
- Tweets12.7K
- Following3.1K
- Followers1.7K
- Likes79.5K
Chairman τao retweeted
Introducing the Bittensor <> Ethereum bridge on Taoswap
You can now bridge USDC or ETH from Ethereum into TAO or any subnet token on Bittensor, and convert TAO back to USDC or ETH.
Powered by @taodotcom.
Chairman τao retweeted
Thank you @GeneticsSociety.
It was a privilege to be in Montreal and share how @opentensor 's Bittensor mission enables Minos: decentralized compute and continuous validation so genomic AI can be trained, benchmarked, and improved.
To everyone who asked hard questions or grabbed merch, thank you. Excited for what's next.
Chairman τao retweeted
PersistentAI is becoming a network operator on UR (@urnetwork).
What we’re cooking:
Joint privacy research - looking into zkAPI & private, open-model routing.
Persistent’s Private Inference (@P_Inference) hooking into UR.io to make it’s traffic anonymous via UR’s proxy network.
Eventual Private Inference subscription layer available to all @urnetwork users.
Say it with us: Private is Better
Turning raw commodities into enterprise ready services.
@KubeTEEAI
bittensor:native
Chairman τao retweeted
Bittensor neoclouds / inference subnets positioned VERY well for this.
Pretty much.
As of the past few weeks, debt markets have almost entirely shut off access to money for any company that doesn’t have substantial revenue independent of AI data centers.
SpaceX closed $40B of debt for GPUs, but we’re investment grade and have Starlink & launch revenue.
That means the 200+ neoclouds are going to see guppy-level death and consolidation. Some will make it through, but not many.
Chairman τao retweeted
Welp. There goes that man again. SN90 @KubeTEEAI
@markjeffrey @ShizzyUnchained @JesusMartinez @jollygreenmoney
SN90 @KubeTEEAI could probably stand to do a little better self promotion. But those who have been watching know that @pamanseau has been delivering steadily with meaningful contributions to the ecosystem.
@ShizzyUnchained @markjeffrey @JesusMartinez @AltcoinMillie
Chairman τao retweeted
The Bittensor $TAO ecosystem weekly update Vol.32
Never a dull moment in the bittensor ecosysytem
A 🧵
Chairman τao retweeted
spooky season
this is the output of our first world model that will be trained by our miners
llms understood text and ate software. world models will understand the physical world and eat operations.
billions of cameras are already installed. almost none of them understand what they're seeing.
most vision ai treats detection and segmentation as the end game. "there's a truck." so what?
for us they're just the eyes and sensors. the world model is the brain. it knows the delivery is late, the bay is blocked and who needs to know.
detection alone means retraining for every new client. that's a services company, not a product. a world model learns how the physical world works once, then applies it everywhere.
and it makes the eyes sharper. a world model can generate endless realistic footage of any site, any angle, any weather, any rare event you'd wait years to catch on camera. fires, break-ins, accidents, on demand. detectors trained on that see things no real dataset could ever show them.
eyes to see, a brain to understand, act and keep training the eyes.
every company will need this to make their operations autonomous. we're building it.
Chairman τao retweeted
We don’t need free movement, we don’t need immigration, we don’t need endless third world migrants.
We need to get healthy British men and women currently claiming benefits back into work. Millions of them.
It is that simple.
Chairman τao retweeted
DeepSeek v4.1 flash 1017 supports image input now
Chairman τao retweeted
For those who understand what @theminos_ai is building on $TAO subnet 107, @jvisserlabs healthcare thesis highlights a huge opportunity. Genomic AI needs reliable tests. HelixForge creates test data with known answers. The investment case strengthens when recurring customer revenue translates into demand for alpha.
Will Eli Lilly become the largest company in the world within 5 years?
@jvisserlabs thinks so.
"What can't be disrupted in the next 5 years is helping people get better."
"The reason $NVDA has seen its multiple go down to me is because it will eventually be disrupted in some way. In the next 5 years? I don't know, but the market is building in that it will be disrupted."
Do you agree with Jordi?
Chairman τao retweeted
This is your children's debt and therefore their future.
If you continue to consent to this, you are consenting to a shit future for them and thus you are a fucking coward.
The TaxPayers’ Alliance revealed this week that public sector debt has breached £3 trillion.
But that is only part of the story.
What the Government is not acknowledging is the vast mountain of additional liabilities it has accumulated, obligations that will ultimately have to be paid.
The real national debt is far worse, standing at £11.7 trillion.
Chairman τao retweeted
SN90 @KubeTEEAI could probably stand to do a little better self promotion. But those who have been watching know that @pamanseau has been delivering steadily with meaningful contributions to the ecosystem.
@ShizzyUnchained @markjeffrey @JesusMartinez @AltcoinMillie
Chairman τao retweeted
The gamma is gonna be flying off the shelf $TAO
A question we often hear from teams evaluating the IOTA SDK: if I train this again, do I get the same model?
We ran 7 independent 16B pretraining runs, ~140 GPUs each. Between them, 585 worker crashes, preemptions and rejoins. The loss curves are almost identical.
Reproducibility is what turns a result into science instead of a one-off. It's also what makes a training run worth paying for.
Most distributed training projects overfit to a single experiment: one model, one recipe, one demo run. The IOTA SDK is general purpose, fault-tolerant and reproducible.
iota.macrocosmos.ai/sdk
Chairman τao retweeted
Async, fault tolerant and distributed by design.
This is liquid training.
A question we often hear from teams evaluating the IOTA SDK: if I train this again, do I get the same model?
We ran 7 independent 16B pretraining runs, ~140 GPUs each. Between them, 585 worker crashes, preemptions and rejoins. The loss curves are almost identical.
Reproducibility is what turns a result into science instead of a one-off. It's also what makes a training run worth paying for.
Most distributed training projects overfit to a single experiment: one model, one recipe, one demo run. The IOTA SDK is general purpose, fault-tolerant and reproducible.
iota.macrocosmos.ai/sdk
Chairman τao retweeted
As frontier labs subsidies going down, majority of companies will be forced to pivot to open weight models that will be 70-80% cheaper for 95% of the performance
Inference is like energy, study Jevon paradox.
Use @engyai
Chairman τao retweeted
I hope the inference optimisation subnets on bittensor:native will compete
@Pareton_ai @cacheon_ai , its time to show your worth!
Chairman τao retweeted
For those who left TAO and/or were once curious about it and moved on to other things, now is probably a good time to re underwrite.
NFA.
bittensor:native
its been a good week being a citizen of the Bittensor network state. here's everything that happened this week
• @lium_io now hosts $30M in GPUs (91% utilization), including 225 B300s. At a $21M run rate
• @affine_io dropped a video model thats ~10x cheaper than the base model with virtually no drop in output quality after 3 days of model optimization
• @engyai hits a new ATH, with 370B tokens processed this week
• @say_gm_ had its highest traffic week, crossing $1M in total volume since July
• @cacheon_ai adds Meta's Muse Spark Open Kernel challenge to their mining competition
• Ingot released Ingot-I quantized Qwen 27B into FP8 better than Alibaba's own quantization in just 1 week
• @trishoolai released HaloGuard EU, an API product that enables companies to be in compliance with the EU's AI act. It's 200x cheaper than Jev
• @yugosai announces their subnet pivot to an intelligent model router
• @metanova_labs subnet-derived compounds underwent wet lab testing and compared favorably with reference compounds including approved drugs
• More subnets entering the buyback game this week with @VantaTrading @ $102K, @Bitcast_network @ $93K, @almnc_ai @ $45K, and @_redteam_ @ $30K.
• eTAO upgrade was implemented into the chain, giving subnet owners control over how they set their emissions. Huge quality of life improvment
Chairman τao retweeted
If this week’s cyberattacks have shown us anything, it’s how absolutely necessary it is for the UK to have sovereign AI capabilities.
This is now urgent.
In the past, the most sophisticated hacking and cyberattack work required considerable expertise, time, and resources. Nation states, intelligence agencies and rogue hacking groups had a substantial advantage - a tiny, tiny number of people.
Today, AI is able to rapidly identify these vulnerabilities, so it's much easier for hostile actors to point an AI at a certain piece of software and ask it to find potential exploits - including in the software that runs many of our key institutions across the public and private sectors.
It is absolutely conceivable that Britain’s hospitals, defence infrastructure, and national grid could be targeted in future cyberattacks. It's already being planned. If these were to be damaged or disabled as a result, thousands could die. If not more.
It is therefore essential that we have the ability to defend ourselves - and the only way to do that is by developing AI sovereignty.
Cybersecurity, and actually much of the global economy, will increasingly rely on AI. It will be an essential tool in combating these sophisticated cyberattacks, alongside skilled people and secure domestic infrastructure.
As things stand, the dominant players are the US and China, but Britain is well positioned to become a serious third force in AI development.
Outside the EU, we have the freedom to set our own domestic regulatory framework, and we could have some of the cheapest energy in the world if we implemented Restore Britain’s energy security strategy.
We cannot afford to be beholden to a foreign power for any of the essentials of life.
Food, energy, defence and, increasingly, AI.
We must build our own infrastructure and train our own models. We must equip our people with the skills to build this industry. Electricians, plumbers, construction workers, and software engineers.
Done right, it can bring benefits to our people, but more importantly - the steps outlined here are absolutely essential for our national security.