Why $ust's peg has everything to do with @CurveFinance's #3pool A thread on $ust's collapse with @dicksonlai_ @GabrielGFoo @themlpx 🧵 @TheSpartanGroup @nansen_ai #luna

https://t.co/cbTeDSm0Za

We all know that insti capital is going to come onchain at scale. In fact, a lot of it already has. Hence, I wanted to map out where said institutional capital will flow into in the next 1-2 years by digging deep into the onchain lending landscape. A TLDR of what I found: 78%

https://t.co/ybASdQmpaj

Worked really hard on this one with @zerokn0wledge_ @Moomsxxx @Louround_ , and found something pretty wild to me: @HyperliquidX is currently doing ~$8.49B in daily volume (as of the time of publication). That's 64% of the entire decentralized derivs market. They're

Why do we think @HyperliquidX could reach $120B FDV within one year? We thoroughly examined the data, crunched the numbers, and built a valuation framework around Hyperliquid's "Everything Exchange" narrative. All details in our latest report ↓

29/ In our space, where narratives move capital faster than code, our information systems must evolve to reward veracity, not just velocity In its current state, Kaito behaves as a mere amplifier for consensus, not an arbiter of signal - imo, a huge waste of its potential

28/ To wrap this up, I truly believe that Kaito has the infrastructure to become a super valuable epistemic (or, infofi) layer in Web3. But when systems of discovery become systems of persuasion, we must question what they amplify, and what they suppress

27/ 5. Systemic yaps reform - Impose disclosure-disclaimers on sponsored content - Incentivize balanced coverage (bull/bear theses) - Introduce scoring for creator integrity: rewarding those who surface risks yappers should be curators of information, not mere hype-amplifiers

26/ 4. Have hype-saturation filters - Rate-limit recycled narratives (ie. “next Solana,” “real yield,” “airdrop farming”) - Penalize trending rankings for repeated memetic-phrases without new substance - Introduce cooldown periods for re-trending without on-chain change

25/ 3. Creating an onchain fundamentals layer This layer should track: - GitHub commit activity - Unique wallet growth - TVL expansion - Verified app usage - Gas costs/contract calls I know these ain't perfect, but we could use these to weigh hype against actual traction

24/ 2. Developing a token-risk dashboard - Visualize token unlocks - Expose whale wallet concentration - Flag protocols with opaque treasuries - Cross-reference smart contract audits Essentially, basic forensic UX for all trending tokens.

23/ 1. Developing a contrarian signal index - Highlight dissenting analysis with real traction (quote tweets, bookmarks, comment debates etc.) - Integrate "engagement-adjusted critique" into trending score - Allow users to filter by divergence from consensus

22/ Once again, I am prolly 1/1000 the builder that anyone at Kaito is, but to me there can be some value in an outsider brainstorming some potential areas that can be substantively improved, so here goes:

21/ Each case follows the same lifecycle: - yapper content begins hype - Kaito rankings surface token - Summaries and alerts spread - Retail buys in at late stage - Insiders/VCs exit - Price collapses - Kaito de-ranks the token The correction always comes after damage is done

20/ 5. Reflexive feedback loop (as aforementioned): - Trending → exposure → mentions → trending again - Retail enters late; Insiders often exit early Because of all the above, the system ends up being epistemically flat, where all mentions count equally

19/ 4. No filters for "fundamentals": - There is no weighting for TVL, GitHub commits, audits, user activity etc. (am sure there are way more substantive/meaningful metrics) - Rug projects can trend just as easily as actual ones, given enough resources and time + effort

18/ 3. Social echo chamber: - Mentions from known influencers, Twitter loops, and Discord chatter are favored - Dissent (and even nuanced critique + dialectical discussions) is systematically underrepresented

17/ 2. Recency bias: - Only recent mentions (1h, 24h, 7d) are tracked - Long-term builders with steady growth are naturally and inherently deprioritized.

16/ These patterns of harm are imo down to 5 structural flaws embedded in Kaito’s architecture that impacts info quality: 1. Positive sentiment bias: - Kaito's summaries and rankings favor bullish phrasing - Critical insights are buried unless they achieve viral traction.