Keerthi Nelaturu

[ricoflan @ricoflan]

The SPAM proposal situation in the Cosmos ecosystem is really a pain this week. Is this symptomatic of bad actors targeting the ecosystem in light of low turn out and protocol vulnerability to voter apathy?? Good stats courtesy of @Plebis_gov

[plebis_gov @Plebis_gov]

This one's from the archive a few years back, but still worth a watch.

@posthuman_validator breaking down his angle on the @cosmos hub governance challenge.

70-75% of delegators don't vote.

Validators end up with power they weren't designed to hold.

Governance becomes backroom politics.

He presents a structural fix: separate tokens for economics, governance, and security.

At Plebis, we are more about the data than aligning ourselves with particular solutions. Regardless of the solution, we feel that better intelligence is a cornerstone of better governance.

One thing this all proves --- two years after his video first posted --- we're all still trying to get this governance thing worked out.

The best is yet to come.

(link to video in comments)

[Shashank @sh1sh1nk]

first post on substack, check it out

https://t.co/R2ItlnGxAa

[plebis_gov @Plebis_gov]

🚨PLEBIS IS NOW LIVE FOR EVERYONE 🚨

🔗 Blockchain Governance is broken🔗

Not because the mechanisms don't work. Because there's simply too much noise. Too much confusion.

Proposals scatter across chains. Voting records vanish into block explorers. Communities make decisions worth millions with little to know context or understanding.

We built Plebis to fix that.

🚪Today we're opening the doors. MVP beta is live. 🚪

Plebis is a governance intelligence platform.

One place to track, analyze, and understand on-chain governance. Our MVP ecosystems include @cosmos Hub, @osmosis , @injective , @axelar , @agoric , @Stargaze.

What's inside:

🔥 Circus Maximus: compare governance activity across every chain, side by side
🔥 The Library: 2,400+ proposals, fully searchable, with complete voting histories
🔥 The Oracle: an AI assistant that summarizes proposals, answers questions, and surfaces what matters
🔥Proprietary indices tracking ecosystem sentiment, staking activity, turnout, and governance health
🔥Validator intelligence with our own grading system
🔥 Leaderboards ranking the most active governance participants
🔥Vox Populi: community sentiment, aggregated and analyzed

This isn't a voting tool. This is the data layer governance has been missing.

Our position on AI is simple: it should inform your decisions, never make them for you. The Oracle clarifies. You decide.

Link in bio.

Come see what governance looks like when you can actually see it.

#blockchain #governance

[plebis_gov @Plebis_gov]

Imagine waking up to find governance participants just voted to fork your token. Liquidity halved. While you slept.

This is why participation matters.

If governance feels too complex to follow, that's exactly what Plebis is built for.
https://t.co/IEhxaI00RX

[plebis_gov @Plebis_gov]

🚀Today, we opened the Plebis ecosystem to the first cohort of Beta testers!

These first users will play in a governance platform that truly didn't exist until today.

✅ AI proposal summaries
✅ AI summary of community discussions
✅ Validator and Constituent Ranks
✅ Ecosystem Comparables
✅ And much more.

There's still time to get into the beta program. Head over to the link in our bio and join the waitlist.

[plebis_gov @Plebis_gov]

Not so fast @VitalikButerin

An AI Proxy for governance is not the solution we need.

ICYMI - Vitalik posted (https://t.co/jaPcZhPT5F) last week, identifying something the blockchain governance community has known for a long time and something that we at Plebis hold as one of the cornerstone challenges that we are trying to solve for: participation is broken.

Too many proposals, too little time, too much complexity. Most token holders don't vote. Those who do often don't fully understand what they're voting on. The result is governance captured by a small, loud minority while everyone else quietly defers. (That's why we built Plebis).

His solution is personal AI agents. Train a model on your past messages, your stated values, and your communication patterns, then let it vote on your behalf. When something seems important or ambiguous, it pauses and asks for clarification. Otherwise, it handles the queue automatically.

It's an interesting idea. It’s an idea that we bandied around for a while when putting our A.I. Oracle together. An idea that we, as well as a large part of the academic community on blockchain governance, think is wrong.

Not wrong in the way bad ideas are wrong. Wrong in the way that a solution can correctly diagnose a problem and still reach a dangerous conclusion.

Vitalik is right about a lot of things here. Human attention is finite. Governance systems that demand constant, expert-level engagement from every participant will always fail. The people who show up consistently are either professionals who are paid to participate or ideologues who have strong reasons to push specific outcomes. Neither group represents the broader community.

AI Delegation doesn't solve this. It just moves the concentration of power one step to the left. Now instead of whales dominating by token weight, you have a smaller group of delegates dominating by accumulated trust. Users click a button once and then effectively leave the room. The delegate remains a permanent proxy until someone changes it. Sure, participation will increase. But the community is no more informed. (Not to mention risks and vulnerabilities that come with handing keys and permissions over to an AI agent).

So yes, the problem is real. Something has to give.

But the answer to "people aren't engaged enough" cannot be "let's remove them from the process entirely."

That's not fixing participation. That's automating the illusion of it.

[plebis_gov @Plebis_gov]

All ecosystems gather in the Circus Maximus.
Your quick glance at what's happening with your favorite project's governance.

🚀Plebis Beta is launching very soon!

@agoric @injective @StargazeZone @osmosis @cosmos @axelar

[plebis_gov @Plebis_gov]

Here we go!

After months of work, the Plebis beta launches in just a few days.

We set out to build something that didn't exist yet: a real governance intelligence platform for blockchain ecosystems.

Not just another place to vote.
A place to actually understand what's happening across chains and make smarter decisions because of it.

What's inside:
✅Unified proposal tracking across multiple chains in one feed.
✅AI-powered summaries and sentiment analysis that cut through the noise.
✅A validator scoring system that shows you who's actually showing up, how they vote, and how consistent they are.
✅Proprietary governance health metrics and cross-chain analytics you literally cannot find anywhere else. ✅Notifications when proposals open, shift, or near closure.
✅A reputation system that rewards the people who do the work.

We're launching with six Cosmos chains (@agoric @osmosis @injective @axelar @cosmos @StargazeZone) and expanding into more ecosystems from there.

The waitlist is live right now at https://t.co/N5Vs7765rX.

Sign up and you'll get access the moment we open the doors. We're talking days, not weeks.

This is just the beginning. Let's go.

[plebis_gov @Plebis_gov]

We've been heads down building Plebis for a while now, and the closed beta launches in just a few days.

Plebis is a governance intelligence platform that will focus on @cosmos as our first ecosystem.

☑️ Track proposals across multiple chains
☑️ Dig into validator voting patterns
☑️ Get AI-powered analysis
☑️ Write Proposals! and more

See how governance health stacks up across the ecosystem.

We're genuinely proud of what we've put together. If you're deep in Cosmos governance (or want to be), we'd love to have you in.

[0tt0 ｜Liberator of Libra (✊☀️) @0x0tt0x0]

Why @0LNetwork is different. They see the red ocean ahead.

[0tt0 ｜Liberator of Libra (✊☀️) @0x0tt0x0]

Q: Can I ask why you remain committed to the industry?

A: I'm not. I can't defend other projects.

But a perpetual endowment for public goods is a great idea, and if it works it will be important.

Unsure it can be done better on traditional rails.

[Open Libra @0LNetwork]

1/ You can't jump across the game board, you need to walk from where you are. A world which could produce UBI is not walkable from here.

[contexto 🦞 @getcontexto]

Cooking 🍜

[LATOKEN @Latoken_cex]

🔔 Upcoming Listing: OPEN LIBRA (LBR) on LATOKEN
💲LBR is a bold fork of Facebook’s Libra, reborn as a fiercely decentralized asset and brought to you from a community fed up with crypto’s empty promises. It’s a fully minted, fair-launch digital property — not a token from a VC-backed foundation, but an open asset built for freedom, savings, and long-term transformation.

⚙️ Why $LBR Stands Out
• No Founders, No VCs, No Labs – 100% independent and community-owned.
• Layer-1 Blockchain – Built on a high-performance Diem+Move framework since 2019.
• No ICO, No Pre-mine – Every coin was earned; no money ever raised.
• 95% Committed to First in Last Out (FILO) – Users have pledged, saved, or donated nearly all supply.
• Designed to hold – Open Libra is not chasing trends; it’s building the future of capital.

🧠 Use Case Highlight
$LBR is engineered to be a resilient store of value and a planning tool for generations to come — a dependable asset to fuel societal progress and transformation over the long arc of time.

🔗 Learn More
• Official Website: https://t.co/6wZRA4vXm9
• Explorer: https://t.co/oNLmT9Th2R
• Docs & Code: https://t.co/F85pnwLnH5
• Wallet: Carpe Wallet
• Discord: https://t.co/uozGfgo6lD

[Sam Blackshear @b1ackd0g]

So excited to see this. Try it out, make your Move code even safer!

[ℏεsam @Hesamation]

Google published a 69-page whitepaper on Prompt Engineering and its best practices, a must-read if you are using LLMs in production:
> zero-shot, one-shot, few-shot
> system prompting
> chain-of-thought (CoT)
> ReAct
> code prompting
> best practices

[Luca Beurer-Kellner @lbeurerkellner]

👿 MCP is all fun, until you add this one malicious MCP server and forget about it.

We have discovered a critical flaw in the widely-used Model Context Protocol (MCP) that enables a new form of LLM attack we term 'Tool Poisoning'.

Leaks SSH key, API keys, etc.

Details below 👇

[0tt0 ｜Liberator of Libra (✊☀️) @0x0tt0x0]

Yes, We Took It

We hard-forked Facebook out of Libra.

P.S. All your base are belong to us.

[Daniyal @Danielkz]

@davidmarcus @pmarca @joerogan OpenLibra forked Libra in 2019 and since then is following a more ambitious vision.

So many lessons have been learned since then and so many challenges are still ahead. Our journey still continues.

Wish you success in your journey at Lightspark!