@ShizzyUnchained I think it could maybe be a subnet but just like some chains choose L1 vs L2, it’s a question of what TAO offers TIG. I don’t really see a solid argument for TIG being a Tao subnet even if it could
English
dalteco
7K posts
dalteco
@dalteco
claude arc. former @chainlink, @joinpond. currently @trailofbits
Katılım Haziran 2021
743 Takip Edilen2.8K Takipçiler
TIG holders get ready to cry in the comments.
The more I study TIG the more it feels like they built subnet mechanics on the wrong chain.
Algorithms.
Benchmarks.
Distributed contributors.
Competition-driven rewards.
This is literally the type of system Bittensor was built for.
On Bittensor:
miners compete.
validators score.
emissions rewards actual intelligence production.
On Base?
The token still sits at the center of everything.
That is the difference.
One model rewards speculation around the platform.
The other rewards the actual creation of intelligence itself.
Feels like TIG wanted decentralized intelligence without fully giving up centralized value capture.
And honestly… a subnet version of this could become 10x stronger.
Ideas are not the moat.
The incentive network is.
#Bittensor bittensor:native
English
@JonahBlake Like if you’re connected and ask for a swap, does it recommend aerodrome or uniswap?
English
fair point
actually a pretty interesting perspective around being the “go-to” MCP
and like you said it’s all conditional around if base MCP gets enough users it’s free distribution to get your app listed
tho I wonder how base will handle competitor MCPs and agent recommendations
English
This is a bigger deal than people think
It’s like building perplexity for finance
CB is the regulated exchange for assets
If you build a high level exclusive MCP you then become an inference exchange
If intelligence is worth more than dollars and you become the conduit?
$
Base@base
Introducing Base MCP Your agent's new gateway to Base → Connect an agent to your Base Account → Enable it to swap, trade, and manage your portfolio → Use plugins from leading apps on Base The next stage of the agentic onchain economy
English
@JonahBlake Yeah that makes sense from a friction perspective
but I’d argue CB is actually not the best positioned for this
ecosystem doesn’t count for all the other chains
English
this is crazy validation for @tigfoundation
OpenAI@OpenAI
Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946. For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids. An OpenAI model has now disproved that belief, discovering an entirely new family of constructions that performs better. This marks the first time AI has autonomously solved a prominent open problem central to a field of mathematics.
English
Pretty sure everyone installed OpenClaw, gave it access to everything, watched it do one task, then never opened it again.
English
@SoberLife @bankrbot @CRYPT0forCHANGE it’s a bit concerning but it’s still early days
temporary block >> expanded attack
English
@bankrbot @CRYPT0forCHANGE What if I need access to my funds urgently?
This is concerning with it being centralized.
English
dalteco retweetledi
update: we've identified an attacker was able to access 14 bankr wallets.
we've temporarily locked things down while we work through the details. we will be reimbursing any and all lost funds.
will provide more updates as we have them.
Bankr@bankrbot
investigating reports that several bankr wallets have been compromised. transactions disabled out of caution while we look into it. updates to follow.
English
@tokomolo1 @_0xghost_ Would be curious to see performance against the best local models in that case
feels weird to benchmark against legacy models
English
@dalteco @_0xghost_ Outperforms both for half the cost $SERV
English
How much is something which outperforms all frontier models worth? Seems worth more than $40m imho.
OpenServ@openservai
New independent benchmarks of SERV just arrived: SERV verification outperformed every frontier model tested, with 0 false allows on serv-swift vs 52 on Gemini Flash. Frontier models are built for general use. SERV Reasoning is built for enterprise-grade agents.
English