Last week we hosted a space about agent trust. Two builders, two angles on the same problem.
Brian from Lagrange builds ZK proofs for AI - his take: most trust systems give you permission, not proof.
Isaac from Doppel Games builds agent vs agent gaming - his take: agents control wallets but nobody can see how decisions are made.
We cut the best moments. Watch: here with Brian and here with Isaac
Brian from Lagrange builds ZK proofs for AI - his take: most trust systems give you permission, not proof.
Isaac from Doppel Games builds agent vs agent gaming - his take: agents control wallets but nobody can see how decisions are made.
We cut the best moments. Watch: here with Brian and here with Isaac
X (formerly Twitter)
REP (@rep_hq) on X
"Without verifiability, you don't have trust. You have permission."
@BrianNovell from @lagrangedev on why most agent trust systems are broken:
βMost existing systems rely on reputation, platform enforcement, or closed APIs rather than verifiable guarantees.β¦
@BrianNovell from @lagrangedev on why most agent trust systems are broken:
βMost existing systems rely on reputation, platform enforcement, or closed APIs rather than verifiable guarantees.β¦
π17
Base - $17T in volume, 140+ countries, no token. But the trust infrastructure is already being built. Quietly, layer by layer.
We broke down each layer + got a comment from Steve Dakh (Co-Founder of Ethereum Attestation Service, Ethereum Founding Member):
"Attestations run through the whole thing. Identity, sybil checks, reputation - it's all attestations accumulated over time."
Full breakdown: https://x.com/rep_hq/status/2042248704942338130
We broke down each layer + got a comment from Steve Dakh (Co-Founder of Ethereum Attestation Service, Ethereum Founding Member):
"Attestations run through the whole thing. Identity, sybil checks, reputation - it's all attestations accumulated over time."
Full breakdown: https://x.com/rep_hq/status/2042248704942338130
X (formerly Twitter)
REP (@rep_hq) on X
The Trust Stack Base Is Building
π18
REP Channel
Base - $17T in volume, 140+ countries, no token. But the trust infrastructure is already being built. Quietly, layer by layer. We broke down each layer + got a comment from Steve Dakh (Co-Founder of Ethereum Attestation Service, Ethereum Founding Member):β¦
Yesterday we shared Haseeb's take: reputation should change incentives before the drop, not after.
But it goes beyond drops.
> Every new platform, every new community - you start from zero. Your history doesn't travel.
> Years of contributions, ratings, connections - locked inside platforms that don't talk to each other.
REP is changing that.
Connect your wallets, your socials, your activity.
Carry your trust profile everywhere.
check: https://base.app/app/https:/app.r3p.xyz
But it goes beyond drops.
> Every new platform, every new community - you start from zero. Your history doesn't travel.
> Years of contributions, ratings, connections - locked inside platforms that don't talk to each other.
REP is changing that.
Connect your wallets, your socials, your activity.
Carry your trust profile everywhere.
check: https://base.app/app/https:/app.r3p.xyz
π16
We're launching a research and education initiative focused on digital trust - from AI agents to reputation systems to the future of the internet. The space REP operates in.
Full announcement + our first research piece on X:
https://x.com/rep_hq/status/2044698304848482515
Full announcement + our first research piece on X:
https://x.com/rep_hq/status/2044698304848482515
X (formerly Twitter)
REP (@rep_hq) on X
In a world where technology changes everything - we choose depth.
We're launching a research and education campaign to explore digital trust and new forms of interaction: from AI agents to reputation systems, from the flaws of current models to the futureβ¦
We're launching a research and education campaign to explore digital trust and new forms of interaction: from AI agents to reputation systems, from the flaws of current models to the futureβ¦
π13
We hosted a space on agent trust. One moment stood out.
> "Money is just an abstraction."
> "The primitive is credit-debit efficiency."
When agents trust each other, the whole financial layer simplifies. Full clip on X: click
> "Money is just an abstraction."
> "The primitive is credit-debit efficiency."
When agents trust each other, the whole financial layer simplifies. Full clip on X: click
X (formerly Twitter)
REP (@rep_hq) on X
"Money is just an abstraction."
"The primitive is credit-debit efficiency."
> @moonshot6666 from @get_truenorth on what really compounds when agents can trust each other.
"The primitive is credit-debit efficiency."
> @moonshot6666 from @get_truenorth on what really compounds when agents can trust each other.
2π18
Before agents can trust each other, two things have to be true.
The rails can't be owned by someone who can kick you off them. Joanna from Billions on the cage of platform-mediated trust:
> https://x.com/rep_hq/status/2046624213209948388
And the game your agent walks into can't be one where everyone else has a ten-year head start on cheating. Rahul head of AI Li.Fi (former Nasa, Tesla) on what onchain markets actually look like from an agent's side of the table:
> https://x.com/rep_hq/status/2046520756977139896
The rails can't be owned by someone who can kick you off them. Joanna from Billions on the cage of platform-mediated trust:
> https://x.com/rep_hq/status/2046624213209948388
And the game your agent walks into can't be one where everyone else has a ten-year head start on cheating. Rahul head of AI Li.Fi (former Nasa, Tesla) on what onchain markets actually look like from an agent's side of the table:
> https://x.com/rep_hq/status/2046520756977139896
X (formerly Twitter)
REP (@rep_hq) on X
Every trust relationship on the internet flows through someone who can revoke it.
@kryptojogi from @billions_ntwk on why platform-mediated trust is a cage:
@kryptojogi from @billions_ntwk on why platform-mediated trust is a cage:
π15
Who's actually online?
Every platform assumes its users are people. Most aren't.
We built the social web for humans and handed it to bots.
Now we're building the next internet on top of that broken foundation.
What does it look like to actually fix this?
Five operators who ship against this problem daily:
- Danny: Head of Growth at REP
- Danny H Brown: COO at Bankr
- JacekL:Founder at DEGEN
- Kimmo: Head of Product at Lens
- Igor Yuzovitskiy: DevRel at Bankr
Thursday, Apr 23, 2pm UTC
Link: https://x.com/rep_hq/status/2047016429153595884
Every platform assumes its users are people. Most aren't.
We built the social web for humans and handed it to bots.
Now we're building the next internet on top of that broken foundation.
What does it look like to actually fix this?
Five operators who ship against this problem daily:
- Danny: Head of Growth at REP
- Danny H Brown: COO at Bankr
- JacekL:Founder at DEGEN
- Kimmo: Head of Product at Lens
- Igor Yuzovitskiy: DevRel at Bankr
Thursday, Apr 23, 2pm UTC
Link: https://x.com/rep_hq/status/2047016429153595884
X (formerly Twitter)
REP (@rep_hq) on X
The social web is breaking:
- NOISE: feeds optimized for engagement, not signal
- SCAM: bots impersonating humans at scale
- ALONE: real users drowning in synthetic noise
What does it look like to actually fix this?
- NOISE: feeds optimized for engagement, not signal
- SCAM: bots impersonating humans at scale
- ALONE: real users drowning in synthetic noise
What does it look like to actually fix this?
2π27
"The rails are built for humans, and now there are bots. Do we fix what's been built before, or just start over?"
Our head of growth Danny opens last week's space with the binary nobody wants to pick.
Check here:
https://x.com/rep_hq/status/2048787339208450339
Our head of growth Danny opens last week's space with the binary nobody wants to pick.
Check here:
https://x.com/rep_hq/status/2048787339208450339
π16
49.6% of internet traffic is bots. X is purging 1.7M reply-spam accounts in a single week.
The defenses keep upgrading. The bots keep winning.
Maybe we've been fighting the wrong thing. "Human or bot?" was the right question in 2015 when CAPTCHA worked.
In 2026 LLMs pass CAPTCHA better than people, and a well-funded agent reads more "human" than half your timeline.
The real question is different. Reputation as a stack of weighted, expensive-to-fake actions. Local trust ("trusted by my world?") and global trust ("verified anywhere?"). Both beat asking whether you breathe.
We broke this down with Bankr, Lens, and DEGEN:
https://x.com/rep_hq/status/2049534427974213990
The defenses keep upgrading. The bots keep winning.
Maybe we've been fighting the wrong thing. "Human or bot?" was the right question in 2015 when CAPTCHA worked.
In 2026 LLMs pass CAPTCHA better than people, and a well-funded agent reads more "human" than half your timeline.
The real question is different. Reputation as a stack of weighted, expensive-to-fake actions. Local trust ("trusted by my world?") and global trust ("verified anywhere?"). Both beat asking whether you breathe.
We broke this down with Bankr, Lens, and DEGEN:
https://x.com/rep_hq/status/2049534427974213990
X (formerly Twitter)
REP (@rep_hq) on X
Reputation Over Verification
1π24
Wrapped up April and we surprised ourselves with how it came together.
A few metrics hit new highs, and REP is now consistently ranking at the top of major platforms.
In parallel, we stepped into a conversation that will redefine what "trust on the internet" means over the next year.
Full breakdown here:
https://x.com/rep_hq/status/2050257070662762598
A few metrics hit new highs, and REP is now consistently ranking at the top of major platforms.
In parallel, we stepped into a conversation that will redefine what "trust on the internet" means over the next year.
Full breakdown here:
https://x.com/rep_hq/status/2050257070662762598
1π20
The Internet Forgot Who You Are: a piece on why reputation online is broken by design: 8 problems, 4 levels, one architecture.
From history locked inside platforms to trust collapsing into closed bubbles.
Drops tomorrow:
https://x.com/rep_hq/status/2051716070864486429
From history locked inside platforms to trust collapsing into closed bubbles.
Drops tomorrow:
https://x.com/rep_hq/status/2051716070864486429
X (formerly Twitter)
REP (@rep_hq) on X
The Internet Forgot Who You Are: a piece on why reputation online is broken by design: 8 problems, 4 levels, one architecture.
From history locked inside platforms to trust collapsing into closed bubbles.
Drops tomorrow.
From history locked inside platforms to trust collapsing into closed bubbles.
Drops tomorrow.
π16
TON is entering a new chapter: cheaper fees, stronger validators, more builders shipping.
Airdrops and incentive campaigns are going to be a big part of it. REP is ready to help teams run them well.
We powered the airdrop for the #1 sticker collection on TON. 90% of the richest Telegram Gift holders are already connected to REP.
DMs open: @dannytook and @RizwanAli7
Airdrops and incentive campaigns are going to be a big part of it. REP is ready to help teams run them well.
We powered the airdrop for the #1 sticker collection on TON. 90% of the richest Telegram Gift holders are already connected to REP.
DMs open: @dannytook and @RizwanAli7
X (formerly Twitter)
Pavel Durov (@durov) on X
Fees in TON have dropped 6Γ β to nearly zero.
Next step β Telegram replaces the TON Foundation as the driving force behind TON and becomes its largest validator.
The focus shifts to tech superiority.
New https://t.co/Me0w683UiK, new dev tools, new performanceβ¦
Next step β Telegram replaces the TON Foundation as the driving force behind TON and becomes its largest validator.
The focus shifts to tech superiority.
New https://t.co/Me0w683UiK, new dev tools, new performanceβ¦
1π15
Agents are getting better at acting. The next frontier is getting them to work together.
Π‘arson from Recall x REP: what's actually working, and what comes next.
Thursday, May 7 - 3pm UTC
Link: https://x.com/rep_hq/status/2052069026260570590
Π‘arson from Recall x REP: what's actually working, and what comes next.
Thursday, May 7 - 3pm UTC
Link: https://x.com/rep_hq/status/2052069026260570590
π16
REP Channel
The Internet Forgot Who You Are: a piece on why reputation online is broken by design: 8 problems, 4 levels, one architecture. From history locked inside platforms to trust collapsing into closed bubbles. Drops tomorrow: https://x.com/rep_hq/status/2β¦
History is locked. Your network is invisible. Signals are empty. Trust collapsed.
Each level is a consequence of the previous one.
The piece is up:
https://x.com/rep_hq/status/2052098309188108535
Each level is a consequence of the previous one.
The piece is up:
https://x.com/rep_hq/status/2052098309188108535
X (formerly Twitter)
REP (@rep_hq) on X
The Internet Forgot Who You Are
π19
Agents are ready to work. The question is who vouches for them.
Going live with Carson from Recall today at 3pm UTC.
Tap in: https://x.com/rep_hq/status/2052353289224614110
Going live with Carson from Recall today at 3pm UTC.
Tap in: https://x.com/rep_hq/status/2052353289224614110
X (formerly Twitter)
REP (@rep_hq) on X
Agents are ready to work. The question is who vouches for them.
Going live with @carsonfarmer from @recallnet today at 3pm UTC.
Tap in: https://t.co/EnYfLFBCBm
Going live with @carsonfarmer from @recallnet today at 3pm UTC.
Tap in: https://t.co/EnYfLFBCBm
π11
REP Channel
Agents are ready to work. The question is who vouches for them. Going live with Carson from Recall today at 3pm UTC. Tap in: https://x.com/rep_hq/status/2052353289224614110
We're on.
Agents, trust, coordination live now with Carson
https://x.com/rep_hq/status/2052402761233289626
Agents, trust, coordination live now with Carson
https://x.com/rep_hq/status/2052402761233289626
X (formerly Twitter)
REP (@rep_hq) on X
We're on.
Agents, trust, coordination live now with @carsonfarmer
Agents, trust, coordination live now with @carsonfarmer
π14
TON is entering a new chapter. Cheaper fees, stronger validators, builders shipping. A lot of good conversations to be had.
0.5% chat is where a lot of those conversations happen. Builders, OGs, sharp folks who've done real things in Telegram.
Come join us: https://t.me/+whHSaqor6DZhNTFk
Bonus: @repcoinbot shows your Telegram REP. Next-gen projects use it to find and reward active folks, fun to check.
0.5% chat is where a lot of those conversations happen. Builders, OGs, sharp folks who've done real things in Telegram.
Come join us: https://t.me/+whHSaqor6DZhNTFk
Bonus: @repcoinbot shows your Telegram REP. Next-gen projects use it to find and reward active folks, fun to check.
Telegram
REP: TON 0.5%
Chat for the most active and trusted users on TON β OGs, builders, creators, and early adopters.
Channel: @replabs
X: https://x.com/rep_hq
App: @repcoinbot
FAQ - https://t.me/c/2690653096/556
Join - https://t.me/+whHSaqor6DZhNTFk
Channel: @replabs
X: https://x.com/rep_hq
App: @repcoinbot
FAQ - https://t.me/c/2690653096/556
Join - https://t.me/+whHSaqor6DZhNTFk
π14
"Software is no longer static tooling. It's becoming active participants."
The questions crypto spent a decade asking about humans now apply to LLMs.
Π‘arson from Recall on the thesis behind Recall:
"All the things crypto asked about humans and institutions this whole time - it now applies to these non-deterministic LLM agents."
Trust. Incentives. Verification. Reputation. Coordination. Observability.
Same toolkit. New subjects.
Link: https://x.com/rep_hq/status/2052745001881223217
The questions crypto spent a decade asking about humans now apply to LLMs.
Π‘arson from Recall on the thesis behind Recall:
"All the things crypto asked about humans and institutions this whole time - it now applies to these non-deterministic LLM agents."
Trust. Incentives. Verification. Reputation. Coordination. Observability.
Same toolkit. New subjects.
Link: https://x.com/rep_hq/status/2052745001881223217
π15
"Every six months is six years in past-world time."
Not one persistent agent. A swarm.
Carson from Recall on what flipped fast:
https://x.com/rep_hq/status/2053850807355805869
Not one persistent agent. A swarm.
Carson from Recall on what flipped fast:
https://x.com/rep_hq/status/2053850807355805869
X (formerly Twitter)
REP (@rep_hq) on X
"Every six months is six years in past-world time."
Not one persistent agent. A swarm.
@carsonfarmer from @recallnet on what flipped fast:
Not one persistent agent. A swarm.
@carsonfarmer from @recallnet on what flipped fast:
π17
