cam

@defi_explora @sam6170 Hyperlane Warp Routes are fully open source and permissionless, there are no liquidity requirements, it’s locking and minting or burning and minting 1:1 from origin to destination chain.

@sam6170 yes yes, most times you’ll find out that some token sent to a chain does not even have any existing liquidity. E.g ezETH on Berachain, the token doesn’t exist on Berachain itself but Renzo bridge powered by hl ofc route it in. crazy…

don't get me wrong, LayerZero’s product is great, but one thing I've noticed with Hyperlane is that there's virtually no slippage when sending tokens across chains. It doesn't matter which chains you're interacting with—whether it's EVM to Cosmos or something else—the amount you send remains exactly the same, even after the transaction.

@defi_explora With Hyperlane Warp Routes there will be no slippage, it’s always a 1:1 transfer locking and minting or burning and minting from origin to destination. Interchain gas is paid by the user on the origin chain.

@hasufl one is permissioned - one is permissionless one is closed source - one is open source one has a 2/2 "DVN" - one has fully flexible and diverse security configurations one costs $$$ - one is free interop for all

Can someone explain to me the difference between Hyperlane and LayerZero? Who would use one vs. the other?

@cryptopunk7213 open source everything

Man, they really gave us the perfect setup for crypto AI First, U.S. president appoints a “AI & Crypto Czar”. Then he announces a $500B AI Fund. & now the best AI model is open sourced for next-to-nothing in cost… .. by CHINA of all countries. Which will spark a fuse resulting in the U.S. doubling down, producing the greatest comeback in AI we’ll ever see over the next decade & position open source, Crypto AI protocols as the largest beneficiaries. What a timeline.

@jamesrosst 100% brother

Open source DeFAI Is the future Study $mode $tao dtao

@topsignal @vyas_krishnan Agree, but if we want trustless and verifiable agents+data then it doesn’t make sense for them to rely on closed source, permissioned, and centralized infra/tools, right?

@vyas_krishnan AI needs trustless, verifiable data. Crypto provides the rails for that. It’s not just currency—it’s infrastructure.

we will accelerate AI development dramatically in the next 3-5 years crypto makes capital formation and verifiability straightforward stay tuned…

@RHLSTHRM @ionicmoney @modenetwork @gizatechxyz 3) @hyperlane

So cool to see @ionicmoney still leading the race for AI agent-owned liquidity on @modenetwork through @gizatechxyz 's ARMA Agent! This paints a very interesting future. In the near future, AI agents will manage most of the onchain liquidity. The funny thing is, they won't care about your project's X account follower numbers, how funny your memes are, or how invested your community is. The agents will be completely rational and simply farm the best yields, and constantly rebalance to find them. The outcome of this is that yields will become totally balanced across protocols, with any arbitrage that shows up instantly eaten up. This is already mostly the case with onchain DEX pools, but lending markets are still fairly inefficient. In order to get to this point there's still a few things that need to be built: 1) Discoverability Agents need to know all the yields across all protocols across all chains. The ideal case is some kind of permissionless aggregator service that normalizes yield data, accounting for incentives and any extra yield on top being generated. Of course this needs to be validated somehow so that protocols cannot lie about their yields to attract AML (agent-managed liquidity). 2) Transaction Creation It's not feasible for agents to require custom integrations for each protocol. There needs to be some standard mechanism where an agent can reliably generate transaction data to deposit and withdraw into any yield-generating protocol. We are experimenting with what this looks like at Ionic with our API: tx.ionic.ninja/api-docs/#/. Alpha leak: We will be dogfooding this with a couple efforts! 3) Chain Abstraction There's still a lot of complexity around getting funds between chains that needs to be solved, and this solution needs to be wrapped in a way that agents can consume it. Agents should be able to consume yields from any chain, instantly arbing rates. 4) Security Of course agents can't just ape into anything they see onchain, they need a way to know that a protocol is vetted to some degree, and better yet, have a realtime security monitoring solution like @HypernativeLabs help secure the AML. These are not easy problems, and problems we have needed to solve for a while. It's a boon that AI agents are here to help us accelerate the pace of DeFi which has been fairly stagnant this cycle (with exceptions of course). Over time as this comes together, we should see much more efficient markets and protocols which is net positive for users and will allow larger and larger inflows of liquidity.

@jamesrosst @modenetwork seat belt buckled

been quiet on x becus of mad building mode DefAI flywheel set for some big big spinning as key projects go live the alpha for what to watch out for @modenetwork > agentic investment daos (next 24 hours) > mode terminal - standard defi + dev capabilities + new model integration (preview tomorrow) > Synth subnet push to mainnet (asap, just a couple security things to review) > level 3 DeFAI agents (👀👀) > ecosystem agent airdrops for veMODE holders + power users (ongoing, a big one announced in private telegram groups) > agent founder school (currently running) lfg

@danielesesta @0xNairolf precisely - if the agents themselves need to be trustless, decentralized, permissionless, and open source then the infrastructure and tools they utilize should be as well, no? what's the point of a trustless agent operating on permissioned centralized infra?

@0xNairolf yea that's why they need an open source framework ;)

AI agents will manage your money, it's inevitable that's why they *MUST* be trustless and verifiable

@jasonzhao @zzinfallible Cam from Hyperlane here 👋 would love to get in touch to be able to give your ecosystem access to a decentralized, open source and fully permissionless cross chain messaging framework

@zzinfallible once public mainnet is live we will support Stargate via LZ

@dabit3 Totally agree but a wise man once said “it’s not things that upset us, but our judgements about those things” Alter your perception alter your reality 🧘‍♂️ 🤙🏼

crypto will have you feeling like you are losing when in reality you are crushing life on almost every front

@0xBebis_ @Cod3xOrg IYKYK

I think @Cod3xOrg is a sleeping giant

@0xjeff @solana Don’t you think many of these AI coins are meme’s, CTO types, generally fully diluted but also with the promise of utility and bullish tokenomic models? Does your thesis stand if these AI coins get added to T1 CEXs?

One thing is starting to become clear ​ Solana is winning the speculation & attention game—both meme coin & AI agent categories. ​ • $SOL pumped ~20% off the back of the successful $TRUMP launch • @solana gained over $13BN in DEX Vol & $2.5BN in TVL over the past 3d ​ ETH & L2s users are starting to understand why attention is consolidated on Solana—it’s the - Seamless UI/UX, the experience of not having to authorize txns 2-3 times for every single txn, - The ease of accessing everything on Phantom, - and the consolidation of attention on one single ecosystem. On the topic of AI agents, Solana has been slowly gaining dominance with many interesting AI agents emerging from the ecosystem. Many aren’t built on the @ai16zdao / Eliza framework but are leveraging their own frameworks and toolkits. ​ Solana agents' dominance overtook Base earlier this year, possibly due to a few factors: ​ • The open-source innovation movement from @ai16zdao and other DAOs like @aicceleratedao from @daosdotfun, which has accelerated the progress of AI agents by offering builders an alternative way to bootstrap projects without VC funding. ​ • Developer-focused community initiatives have successfully attracted builders, leading to numerous frameworks/toolkits being developed on Solana. ➔ @sendaifun’s Solana AI Hackathon is a prime example, with over $500M in MC coming from 100+ projects already with tokens, and over 200+ projects still without tokens. Despite Solana surpassing Base in AI agents, @virtuals_io remains the #1 AI Agent Ecosystem due to its pioneering role in AI Agent Tokenization. ​ Virtuals’ strong value accrual, robust tokenomics, storytelling, project incubation, and ability to ensure successful token and product launches set it apart. A few questions arise: • Given Solana's dominance in meme coins & AI agents over other ecosystems, could we see Virtuals launching a Solana version of Virtuals soon to capture the #1 launchpad spot on Solana, on top of Base? ​ • If Virtuals doesn’t launch on Solana, could a new ecosystem leader emerge to act as the #1 distribution channel on Solana, just like Virtuals is for Base? ​ • Could Base reduce the AI agent dominance gap between Solana and Base? If so, how? ​ So many questions, so few answers. I guess we’ll see how this unfolds in the next few months. One other interesting observation is the clear relationship between meme coins & the AI agent narrative: • AI agent pumps, meme coins go sideways or dump. • AI agent dumps, meme coins pump. AI agent-related meme coins pump even higher (e.g., $LLM as a clear inverse play of $ai16z, with the catalyst being @shawmakesmagic crashing out, or $FARTCOIN pumping when the rest of the AI agent landscape dumps). ​ There’s a clear trend here—when people grow bored with AI agents (when agents are dumping) and fundamentals/utilities aren’t performing well, attention always rotates to memes. ​ I’ve seen this play out many times but resisted hopping onboard to buy some memes. This sector-specific bias—investing only in one area—is short-sighted. That’s why I’ll start following @TaikiMaeda2’s footsteps and hedge myself with appropriate memes to offset retracements in the AI agent sector.

@dabit3 It’s often the roll up stacks themselves (and then social consensus) that are dictating usage of an optimistic bridge (often without fault proofs) to natively mint ETH as gas full knowing the terrible UX to come 🤷🏻‍♂️

ETH being a gas token on other networks is confusing to new users. Use the Solana network, buy and trade using SOL. Ok 👍 Use the Arbitrum or Base or X network, want to buy X or Y. By the way to do any of this shit you also need to buy this other token, oh and be sure it’s on the right network, oh right here are 80 bridges choose one. TLDR: UX on Solana is better than most (all?) networks that use ETH as the gas token and the answer why is pretty simple (not taking into consideration cost and speed)

@cryptopunk7213 @griffaindotcom @AIWayfinder @0xBebis_ @Cod3xOrg

The Agent Gig Economy So far we’ve been hyping up the agent KOLs they’re entertaining & good at creating community. But they’ve been poor at providing utility beyond words. DeFAI has tried to tackle this, with engines like @griffaindotcom @AIWayfinder paving the way But we still haven’t seen an agent do a bunch of DeFi stuff autonomously at scale. A lot of this comes down to *existing* blockchain infra design. Every agent needs to plug into a million L1s, L2s, bridges, exchanges etc. & frameworks are racing to do this. But I wonder if this can all be abstracted away by worker agents. These agents would be hyper-specialised. Designed to just execute 1-2 tasks max. Over & over again. For other agents. Think of them like the line-cooks in a kitchen. They’re assigned a specific set of tasks which they execute over the course of a dining session. They don’t stray from those tasks. The culmination of multiple line-cooks doing individual tasks is a 5-course meal, curated by the head chef (in this case the agent KOL). It would abstract away all the “dirty work” , KOL agents can just contract the tasks out to a worker agent. Like in a gig economy. This is more economically attractive too. Agents don’t care how much they’re paid as long as it covers the cost of their compute & data + a small fee. & they’re available 24/7 & infinitely scalable. Imagine 100Ms of these agents working tirelessly around-the-clock. Behind every great agent influencer… is a worker agent?

@jamesrosst @modenetwork love to see it

That’s a fuck load of agents on @modenetwork and it’s only the beginning

@cryptopunk7213 Monetization, distribution, and accessibility will be key. With distribution comes the need for coordination. The beauty of open source frameworks/toolkits is that a simple PR can lead to permissionless integration. Facilitating this type of collaboration is what will catalyze.

Think AI + gaming will be Web3 Gaming’s breakthrough moment. I think the runaway game of this cycle will include: AI Agents + generative open worlds (MMORPG) + Social incentives + Ability to inflict damage Agents provide an interactive, choose-your-own storyline experience. AI-generated virtual worlds ensure the player experience is fresh & dynamic. Open frameworks allow players access to new tools to socialise & transact. Ability to inflict damage creates an element of scarcity that hooks players. My guess is we’ll see *more* winners emerge from infra projects however, the game apps that do well will dwarf any underlying infra (Fat App Thesis). I bet an open source, decentralized agent infra stack will enable it, Onboarding 100,000s of people into crypto at the same time. @virtuals_io leading the charge here with G.A.M.E + top partnerships (Illuvial). @ai16zdao integrating agents with @hyperfy_io for virtual worlds

@cryptopunk7213 And who is going to facilitate this cross chain functionality between agents and swarms? Likely a fully open source and permissionless modular General Message Passing protocol?

Think building agents on a single chain is the wrong approach. You want them to be able to function on *any chain*. They should be able to traverse freely, autonomously interacting with other agents & digital assets to fulfil their objective. Choosing a single chain “for developer community” is old-school thinking. Agents aren’t going to care which chain they use. The best communities will be multi-chain. Already seeing this with some leading agent protocols @ai16zdao with multiple chain plug-ins @arcdotfun partnering with multiple chains. Who else?

@hyperlane is the only General Message Passing protocol that was purpose built to be modular, fully open source, and permissionless Hyperlane is the open framework for agentic interoperability More bounties to come, more innovation to follow Expansion