The Attention Capital Markets Thesis
What’s Next for Memecoins: Why Attention Matters, AI, and the Rise of Speculative Consumer Apps on Solana.
The ‘Attention Capital Markets’ is a niche term, loosely derived from Solana’s ‘Internet Capital Markets.’
While Internet Capital Markets are based on the idea that anyone can get listed, i.e, launch a token on the internet (instead of traditional gated capital markets like NYSE). Attention Capital Markets is all about monetising or tokenising attention. I’ve always believed that memecoins are ‘attention assets. ’
Why?
Prices and volumes are highly correlated with attention — i.e., if any particular meme, event, or tech narrative has insane attention, the number will go up.
Attention Theory: Valuation = Fundamentals + Attention
For memecoins, it’s all about attention, while for stocks like NVIDIA, it’s a combination of fundamentals (based on revenue) and attention (broader AI and GPU shortage narrative).
Attention Economy [Social Media]:
We’re living in an attention Economy. Major platforms like Google/Meta are famously called “Merchants of Attention,” as they primarily trade our attention and monetize through ads.
More Attention = More Platform Revenue
Thanks to the rise of short-form platforms like TikTok and Instagram, attention spans have significantly declined — you’ve now <3 secs to capture attention as users seek instant gratification. With the rise of AI, where the cost of digital production reduces significantly — from code to content — attention becomes even more scarce.
Memecoins — Speculate Attention
Memecoins are the ‘TikToks’ of financial assets. Thanks to memecoins (e.g., on Solana), one can speculate on what will be relevant or fun, and for how long, to capture significant attention. The magnitude of attention is simply the market cap of a memecoin. This attention currently comes primarily from crypto and adjacent audiences, which will keep expanding to broader audiences. Everyone trading memecoins is playing the game of “Attention Arbitrage,” i.e, — buying what’s gaining attention and selling what’s already had its moment. For instance, if you had to bet on the attention of ‘Crypto x AI’ — you’ll rotate your political coin portfolio to a few AI coins. While crypto/Solana enables the creation of new markets and the trading of attention, on-chain markets remain relatively inefficient due to:
- Insiders/Cabals (Information asymmetry)
- Mechanism Failures (Improper market structure)
These inefficiencies result in deviations between ‘true attention’ and ‘market prices.’
Declining Attention Spans:
Just like with short-form content on social media, attention spans for memecoins have been shrinking rapidly.
Previously, peak billion-dollar memecoins had lifecycles of 2–3 years or months (e.g., Dogwifhat). Now, they last just 2–3 hours (e.g., Libra) — all this rapidly evolved within the span of a single year.
Declining attention spans in crypto → Higher speculation velocity.
And it’s not just memecoins — the duration of metas in crypto (think AI Agents meta or the recent SocialFi wave) is getting shorter and more speculative by the day. Declining attention spans are also part of why we’ve seen weaker alts this cycle — capital and attention are both finite. Rotation from one coin or meta to another is faster than ever.
~40k tokens are still being launched daily only on one chain — Solana. Thanks to memecoin mania, we’re at an inflection point.
There’s no reason why it won’t go exponential from here. Think of it as a “Youtube moment” — when anyone could have published on the internet permissionlessly (and all media broadcast regulators lost control). We are at a similar cusp on tokens (or financial instruments), where now anyone can launch a token with just 2 clicks.
Implication of the Attention Capital Markets:
1. More Speculative Consumer Apps Using Attention Assets.
— Pump: @pumpdotfun is the prime example — an infinite content flywheel where any kind of attention can be monetized in under $2 and less than 2 minutes. The live stream feature was an early glimpse into attention capital markets, where visual attention could be directly monetized (might be launching again with moderation soon 👀)
— SocialFi is also an early glimpse. While most SocialFi apps have failed to be long-term sustainable due to misaligned structures — @timedotfun is one of the best examples of Attention Capital Markets. It lets creator directly monetize their time and earn revenues from the trading fees (via LP fees).
For instance, OnlyFans via Attention Capital Markets would be to speculate on models (you discover models early, make bags) — while creators earn revenue from trading fees or their own token allocations.
More Attention = More Volumes/High Prices = More Revenues
— @KaitoAI Yaps is another platform that has quantified ‘quality CT attention’, leading to ‘InfoFi’. InfoFi is using market forces to determine where attention should flow instead of social media algorithms. @wallchain_xyz is another platform building for “AttentionFi”. @noise_xyz will use Kaito’s mindshare data — where anyone can long and short trends (or attention).
@Polymarket and the 2024 US Election are other examples of using markets to determine outcomes.
— Attention Discovery: Copy trading platforms like @VECTORDOTFUN and @gmgnai help in attention discovery via feeds/broadcast. With public information more accessible than ever — especially via advanced reasoning models — alpha or access to private information is becoming increasingly scarce and valuable. Curation will keep getting more challenging.
We’ve barely scratched the surface — and there are hardly any platforms that have escape velocity in terms of both numbers and volume (except Pump).
More Speculative the App → Shorter the Lifecycle of the App
Again, launchpads like Pump are exceptions (due to infinite content flywheel), but generally, all speculative apps go insanely viral for a few days (generate huge revenues) and die more quickly, primarily due to mercenary capital rotation. While novel (or non-speculative) has more staying power, most fail to gather any attention due to their non-speculative nature. Remember, Speculation is a bug, not a feature in Crypto.
What’s Next?
1) Experimentations around Vertical Pump
i.e Pump but with either or both curation and mechanism:
— Niche curations like music, dating, time, and so on. Specific verticals will allow for better feed curation (FYPs) and, hence, better consumer UX. Further, a few more ‘AttentionFi’ products will emerge tapping existing social graphs. For instance, Kaito is tapping the Twitter social graph — mass popular platforms like TikTok, Discord, and YouTube are still underexplored.
— Specific Mechanism: Pump isn’t one-size fit all; we need custom market structures for different use cases. For instance, pre-sale or dutch auction is better suited for project-based token launches. This also involves experimentation around creator-fee sharing, niche AMM design, tokenomics/liquidity locking, pricing curves, and more.
Memecoins (or Attention Assets) are here to stay: Memecoins on Solana were a glimpse of an incredible phenomenon. Yes, like with every market structure, there were insiders, and many tokens reached highly bloated valuations — pumped by cabals. But we can’t discredit the tech (Solana DEXs) that powered $500B+ in volume in just one month!
Cabals and insiders are clear signs of a highly inefficient market — but the best part is, information travels much faster on the internet and on-chain (especially on Solana). Now’s the time to design and experiment with new market structures — pushing us closer to efficient markets.
2) AI, Vibe Coding, and Reducing Cost of Production:
AI significantly reduces the cost of producing apps — which means we’ll see an exponential rise in the number of apps being built. Vibe coding is an early signal: people are already shipping MVPs within hours. Soon, they’ll be building production-level apps just as quickly. Imagine the app store — 1 year from now. Getting attention (and discovery of apps) would become increasingly harder.
The same will eventually be true for crypto apps: I don’t think AI will be writing production-level smart contracts anytime soon — battle-tested and audited contracts will still be preferred. But AI will seamlessly integrate with existing contracts, speeding up the development of crypto apps. While current models don’t have perfect context on crypto SDKs, that’s changing. Soon, AI will be able to vibe code crypto apps just as fast.
Further, MCPs allow them to have crypto context — for instance, @sendaifun integration with @cline helps Cursor/VS code to get Solana SDKs context. Still not perfect, but we’re getting there.
Once building crypto apps becomes easier, we’ll see a wave of speculative apps emerge. People will spin up experimental apps (or games or ‘meme apps’) in just a few hours — launching them with tokens to bootstrap early users and community.
Tokens will give apps the speculative flywheel (token go up → more users → token go up).
A few might generate meaningful revenue ($100K–$500K) before dying out in 4–5 days. Stickiness or retention would become more scarce, which v few apps will nail. Just like the agents meta, every app might launch with a token (h/t my interview with blocmates) — giving rise to a new wave of “app-coins.”
One project I’m watching closely is @devfunpump, which is building at the intersection of vibe coding and tokens (and will soon integrate embedded wallets). As apps become easier to build, taste (or ability to gain attention) becomes the new moat. For instance, AI can generate UI, but the “aha moment” of a delightful UX still needs human taste.
Agents and Apps.
Right now, there are barely ~50 solid crypto mobile apps — most of them wallets. Personally, I’m excited to see more mobile apps in crypto. They have the potential to onboard more users.
The platforms that win will be the ones that help build, launch, and distribute mini-apps or agents — whether in crypto (with tokens + wallets) or beyond.
I’m incredibly excited about how both agents and apps evolve — powered by AI and Attention Capital Markets. 2025 will be all about building to accelerate agents and apps — can’t wait to share what we’ve been cooking.
