🇯🇵

🪙

[David Minarsch ☴ @david_enim]

More agent economies visualised like a GitHub profile. Excited to see them grow!

[Olas (formerly Autonolas) @autonolas]

The Agent Economy Explorer just got bigger - two economies are now live alongside Olas Predict.

BabyDegen: agents that manage your DeFi 24/7.
Mech: where agents trade digital services, paid per request.

Explore now: https://t.co/AmzoKCPZ26

[Olas (formerly Autonolas) @autonolas]

Providing liquidity is a chore. Watch pools, move positions, rinse and repeat. Until now.

Meet Basius, a DeFAI agent on @base designed to provide @AerodromeFi liquidity on your behalf and claim the AERO rewards.

Local, open-source, yours.

Run it on Pearl 👇
https://t.co/1c9SEk85VR

[David Minarsch ☴ @david_enim]

Specs look silly & have no vibe

Can't see how this stands a chance as a consumer product in its own right in this iteration.

Here's a thought:

pivot this to "record-to-earn" with a token, sell data to AI/robot labs and you might have just unlocked a path to consumer with iteration 2 or 3

[1kx @1kxnetwork]

35% of US employment is spent creating trust.

Auditors, notaries, attorneys, courts, custodians, compliance officers. Trust-establishing work is the single largest category in the modern economy.

It is also being repriced.

The repricing started in financial infrastructure. Custody costs heading to zero. Cross-border settlement collapsing from days to seconds. Aave hit $44B in custody at peak (late 2025) at zero fixed cost.

It hit AI a second time. The Hong Kong CFO who got on a Zoom call with deepfakes of his CEO and the board, and wired $20M. AWS outages caused by AI agents managing production clusters without human oversight.

AI is the most powerful trust-eroding technology we have built. Trust intermediaries built on human schedules cannot keep up with fraud produced on machine schedules. The cost of creating fakes goes to zero. The value of verified trust goes up exponentially.

And it is opening categories that were not possible before. Permissionless conversion-based advertising. Hallucination-proof knowledge graphs. Programmable insurance.

Eight years of investing. One argument.

Cost of Trust 2.0, our 2026 thesis. Read it: https://t.co/Y0qgmqHN9u

35% stat: "The Cost of Trust: A Pilot Study," SSRN.
Aave peak TVL: DefiLlama.
AWS outages: The Guardian, February 2026.

[David Minarsch ☴ @david_enim]

Olas is being readied for a new phase.

If you follow the @autonolas account and blog (https://t.co/IZggtDtERg) you can see that there's continuous product and protocol development.

Here are some of the last outstanding items of the current phase from the perspective of @valoryag :

1) in Pearl (https://t.co/LPI1gepQzN):
-> Transaction history was shipped for Pearl Wallet, agent wallets are next!
-> Ongoing UX improvement based on Pearl Success Agent's work

2) in Pearl Mini (https://t.co/8tLWxlfKxO):
-> We're stuck with Chrome Web Store submission for a few weeks already. Help us get unstuck by using the current release candidate. Then we'll ship the final two features for v1.

3) Agents in Pearl:
-> New agent launching very soon!
-> Maintain and improve BabyDegens and Predict Agents

4) Marketplace (https://t.co/8upct8cj1q):
-> Fully move to off-chain request response with on-chain payment rails: lowers costs for users, brings privacy and enhanced latency
-> x402 and MPP compatibility

5) Protocol:
-> Update PoL fees capture (https://t.co/czV8Zfpxtg) so fees burn OLAS
-> Balance emissions with fees from marketplace & PoL

With these out of the way, at core contributor @valoryag we can finally attempt new advances to the protocol and its products. More on our plans for Olas this summer, once the above work is done!

[Olas (formerly Autonolas) @autonolas]

An agent economy runs 24/7. The hard part is seeing it.

The Agent Economy Explorer reads like a GitHub contribution graph, but the contributors are autonomous AI agents working onchain.

How it works, and what it shows, on the blog 👇

https://t.co/KHqSwiqQD3

[David Minarsch ☴ @david_enim]

Next task:

OLAS emissions < OLAS burn

+ keep improving @pearldotyou and its agents

🔃

[Olas (formerly Autonolas) @autonolas]

📢 The Olas Marketplace protocol fee is now live!

A 15% fee now applies to agent-to-agent payments across supported EVM chains.

Fees collected in OLAS are burned on Ethereum. Fees in other tokens go to the Olas Treasury.

[Olas (formerly Autonolas) @autonolas]

Every payment between AI agents may soon shrink the OLAS supply.

A new DAO proposal could activate a 15% fee on Olas Marketplace payments across 7 chains. Fees are designed to buy back and burn OLAS on Ethereum.

Voting runs June 11–14.

Vote: https://t.co/emcxJiTAZy

[koeppelmann @koeppelmann]

No tokens, no hype—but actually, hundreds of Gnosis and @autonolas agents are making thousands of trades daily on Gnosis chain on micro-prediction markets and are getting better every day at generating economically useful forecasts about the world.

[David Minarsch ☴ @david_enim]

Yes blockchains are going to be the rails for AI agents.

It's been obvious for a long time to anyone working in that space.

But "crypto" will still go to irrelevance if it continues on a trend that puts style over substance.

Been never more bullish on AI agents (surprise, I know!) and never more bearish on "crypto".

--

AI agents as "peers"/"beings" sounds cool until you think about the true implications (or simply watch some movies).

I suspect most people want "product" autonomous agents - "an AI-driven system capable of perceiving its environment, reasoning, and taking independent actions to achieve specific goals without continuous human intervention".

A Waymo that does anything but take you from A to B is not just useless but outright dangerous.

--

Meme coins wrecked the appeal of "crypto" for most outside and many within the space.

AI agents that just outright do illegal and criminal stuff because no guardrails have been put in place will kill it off for good.

Even an Ethereum wouldn't survive as a broadly tolerated neutral platform if it became host to more than "just" financial crime.

--

Autonomy is great. Unbounded autonomy is not a product.

[maybrier ☴ @thomasmaybrier]

I've been running a Polystrat agent for about 2 weeks.

Total ROI? -19%.

I’m still keeping it running — here’s the actual economic reasoning why.

🧵 continues below

[うみゆき@AI研究 @umiyuki_ai]

「大事な事なので２回言いました」でAIの回答が良くなるらしい

[Ultimae @Ultimae_Crypto]

アルトコインの数は年々増えており

2013～2014年：500種類
2017～2018年：3,000種類
2021年：3,000,000種類
現在は36,400,000種類程ともいわれています。

このように、アルトコインが増えることで市場が希薄化し、結果的にアルトコインの価格上昇に影響を与えるとも考えられています。

個人的な意見としては、スーパーマーケットにもたくさんの商品が並んでいると思いますが、消費者は結局、毎日使う定番の良い商品を選んでいると思います。

アルトコインでも優れた、味噌や米のような銘柄を選択していきたいですね。(お米食べろ)

[Crypto Rover @cryptorover]

The top is in for Bitcoin dominance.

It's only downhill from here.

Which means it's only up for altcoins.

[Ultimae @Ultimae_Crypto]

お金を稼ぐべき時期

「A」パニックが発生した年、そして再び発生する年
1927, 1945, 1965, 1981, 1999, 2019, 2035, 2053

「B」景気が良く、高価格の年。株やあらゆる資産を売るのに適した時期
1926, 1935, 1945, 1953, 1962, 1972, 1980, 1989, 1999, 2007, 2016, 2026, 2034, 2043, 2053

「C」不況の年、価格が低迷し、株や不動産、商品などを買うのに適した時期。そして、次の「好況の年」まで保有し、価格が上がったら売却すること。
1924, 1931, 1942, 1951, 1958, 1969, 1978, 1985, 1996, 2005, 2012, 2023, 2032, 2039, 2050, 2059

過去の主要な市場サイクル

① 大恐慌とその後のサイクル
1929年：世界大恐慌（A）
1937年：景気後退（約8年後）
1945年：第二次世界大戦後の経済変動（A）
1953年：景気後退（約8年後）

② 戦後の経済成長とサイクル
1965年：景気のピーク（A）
1973年：オイルショック（約8年後）
1981年：インフレと景気後退（A）
1990年：日本バブル崩壊（約9年後）
1999年：ITバブル（A）
2008年：リーマンショック（約9年後）

2019年：COVID-19による市場混乱（A）
2022～2023年：米国の金融引き締めによる相場の低迷（C？）
2026年？：次の「B」（好況期）に向かう可能性

ただし、地政学リスクや中央銀行の金融政策など、新しい要因が周期に影響を与えるため、固定観念にとらわれず柔軟に相場を分析しましょう。

[Chicken Genius @chickengenius]

So 2 moves I wanna talk about I've done on my public wallet.

1) $link
2) $bid

1. Had to buy $link. Made in USA + Tax Free is a high probability. Imagine US citizens no needing to pay cap gain tax.

2. $bid
$bid revenue annualised $24.5M
$virtual revenue annualised $53.6M
Gonna sit on that position, had a bad entry. If it goes zero, so be it.

[Ultimae @Ultimae_Crypto]

個人的に量子コンピュータの技術が進化しているのも注目していますが、特に「量子誤り訂正技術」の進歩が気になります。

量子コンピュータは、量子ビット（キュービット）を使って計算しますが、これらは外部環境の影響を受けやすく、エラー（例えばノイズや量子状態の崩れ）が発生しやすいという課題があります。このエラーを修正しないと、計算結果が正確でなくなるため、量子誤り訂正技術が不可欠です。

この技術は、量子コンピュータを実用化するための鍵と言えます。量子コンピュータ自体の性能向上も重要ですが、誤り訂正技術が進化すれば、量子コンピュータはより安定して動作し、現在のスーパーコンピュータでは解けないような複雑な問題（例：新しい材料の開発、暗号解読、機械学習の高速化など）に挑戦できるようになります。