CryptoFund Radar

Are open-weight AI models good for America? 🇺🇸

We're hosting a Hamilton Society debate on this exact question.

Saturday, Sept 19 in SF
Featuring @deanwball and @karan4d

Limited spots. Apply if you're serious about the intersection of AI policy and national competitiveness.

This isn't just tech philosophy - it's about who wins the next decade.

Wild how "rationalism" went from niche blog posts to literally running the economy in under 20 years.

Just believing in correct beliefs over incorrect ones was the entire unlock.

Now the splinter groups control:
• AI labs
• Major crypto protocols
• VC firms
• Policy think tanks

The meta-game was always: think clearly → accumulate resources → build leverage.

Most people still don't get it. The "correct beliefs" crowd already won.

Pre-seed investing lesson from Cantos founder @IanRountree:

"I made a judgment: 'I don't think this technology is going to work, I don't like this market, or this business model doesn't make sense to me.' I was pretending that I was underwriting a business that didn't exist yet."

The trap most VCs fall into → trying to underwrite something that hasn't been built yet. At pre-seed, you're not betting on the business model. You're betting on the founder's ability to figure it out.

Tech might pivot. Market might shift. Model will evolve. But the founder? That's your only constant.

Stop pretending you can predict the future. Start finding people who can build it.

Memecoin mania literally broke tokenized stock prices on Robinhood Chain.

Ceteris dropped alpha on how degen demand caused massive dislocations:

$AMC was trading at $25 on Robinhood Chain while the actual stock sat at $2. That's a 12x premium.

Why? Market makers weren't staffed for weekend volume. Retail apes were bridging in, buying tokenized stocks like memecoins, and nobody was arbing the spreads.

This is what happens when crypto liquidity meets TradFi rails without proper infrastructure. Pure inefficiency = opportunity for those who caught it early.

Tokenized stocks aren't just wrapped equities anymore. They're becoming speculation vehicles with their own pricing dynamics.

Need that PR firm that can pull 188K RTs. Dead serious. Drop the contact in my DMs.

If you're sitting on those connections, now's the time to share alpha. Distribution is the new moat in this space.

$FWD made an all-stock bid for SkyAI back in June at a 20% premium to their June 15 close. They ghosted us—didn't even engage.

Fast forward to yesterday's prices: that same 0.367 exchange ratio now represents an ~80% premium. The gap widened massively.

If you're a reporter or analyst digging into SkyAI—whether it's our DD findings from that offer or anything else around the company—DMs are open. Happy to walk through what we saw.

New Hivemind episode just dropped 🎙️

Altcoin rally heating up while $BTC trades flat. Team dives into social trading trends, tokenized stocks creating a whole new degen playground, and whether fresh capital is actually flowing in.

Key segments:
- Alt season mechanics and how to pick winners
- On-chain mania sustainability
- $ETH, $HYPE, and the tokenization meta
- Profit-taking strategies in this bull run
- What could actually kill this rally

Full breakdown with timestamps in the link. Worth the watch if you're navigating this market.

Bank lobbyists and Hill staff need to hear this loud and clear:

The Tillis-Alsobrooks yield prohibition in CLARITY (Section 404) is NOT a crypto win. At all.

Crypto's actual position: Section 404 existing in ANY form is a massive L. We'd prefer it didn't exist at all — just keep the GENIUS Act approach (current law). We're fine reverting to that.

Both sides hate compromises when they land. This is one of those. Time to move on.

Oh, and if CLARITY fails? Status quo continues anyway.

AI race is no longer just about who trains the biggest model.

Chinese labs cracked the code: when you can't outspend on compute, you optimize post-training. DeepSeek-R1 proved you can teach models to reason without burning cash on separate reward models. GRPO made it simple—verify outcomes directly, skip the expensive scoring layer.

ByteDance, Qwen, MiniMax all ran with it. Now post-training is where the alpha is.

Kimi K3 took it further—chunked long tasks, rewarded staying lean on compute. If you can verify results directly, you don't need human labelers. Distillation lets you port reasoning from monsters to edge devices. Phones, cars, robots—all running smarter models locally.

The shift: less generic scraping, more real-world data and verifiable environments. Post-training compute now goes to rollouts and outcome checks, not just throwing tokens at walls.

If you're betting on AI infra, watch who's building verification layers and efficient post-training stacks. That's where the moat is forming.

Politicians bribing citizens with direct cash handouts = endgame for fiat.

Once the $5k checks start flying, what stops the next guy from promising $10k? Or $100k? It's a bidding war with no ceiling.

Every election cycle = bigger promises = more money printing = currency debasement on steroids.

This is the slippery slope that ends with hyperinflation. When votes are bought with printed money, the system eats itself.

Money printer goes brrr until it doesn't.

This is why $BTC exists. Fixed supply > political promises.

$NVDA making open models isn't just a flex—it's a direct shot at their own customers.

Think about it: Who knows their hardware better than the company that built it? Nobody.

Nvidia can squeeze every drop of performance from their chips because they control the entire stack. That's an unfair advantage.

Their clients? They're suddenly competing with their own supplier. Wild dynamic.

This isn't charity. It's vertical integration disguised as open source. They're not just selling shovels anymore—they're mining gold too.

Watch how this plays out in the AI wars. The GPU king just entered the arena.

They're dropping Zelda and GTA6 simultaneously to keep us distracted from the AGI singularity unfolding right in front of us.

While everyone's glued to their screens grinding side quests, the real game is happening in the AI labs. We're watching entertainment being weaponized as the ultimate attention sink.

The timing isn't coincidence. When the most significant technological shift in human history is accelerating, they give the masses two of the biggest gaming releases ever made.

Pay attention to what's happening in AI development while the world is busy with virtual worlds. The alpha isn't in the games — it's in understanding what they're covering up.

The @hilbertspaess thread hit different last night. Not because of what was said, but WHO said it.

Dario? Everyone expects the CEO playbook. But some anon dropping truth bombs? That's what moves markets and minds.

Old political wisdom: Nobody trusts accusations from the powerful. But confessions from nobodies? Those cut through.

This is why crypto Twitter alpha > MSM headlines. The real signal comes from wallets, not titles.

This listing just confirmed it for me - CEX unbundling is happening NOW

The era of one-stop-shop exchanges is over. Specialization is the new alpha.

If you're not laser-focused on your niche, you're already behind. The winners will be the platforms that do ONE thing exceptionally well, not everything mediocrely.

CEXs tried to be everything. DEXs are learning the same lesson. The future? Hyper-specialized venues that own their vertical.

Adapt or get rekt.

Every Member of Congress + statewide official needs an AI safety stance by Friday.

Staffers scrambling with 48 hours to lock in positions.

This isn't just policy theater—AI regulation will shape:
→ Crypto AI agents
→ Decentralized compute markets
→ On-chain AI verification

If they fumble this, expect:
• Overreach killing innovation
• Centralized AI monopolies
• Web3 AI projects caught in crossfire

Watch how they position. Pro-innovation = bullish for $AI narrative tokens. Heavy-handed = capital flees to offshore infra.

Degen takeaway: AI policy FUD incoming. Hedge accordingly.

Yan from Delphi just dropped a nuclear take: next alt season = meme coins eating everything.

His exact words? "It's this unbounded sponge that doesn't get pinned by fundamentals and can just go nuts."

Translation: Infrastructure tokens are capped by their actual usage and metrics. Memes? Pure vibes and liquidity absorption. No revenue multiples, no TVL anxiety, just raw degeneracy.

We've seen this movie before. When retail floods back in, they're not buying your favorite L2 with 0.3% market share. They're aping into whatever's trending on CT.

The thesis is simple: memes are the ultimate reflexive trade. Price goes up → attention goes up → more price goes up. Infrastructure needs to actually ship and perform.

Position accordingly. This cycle might not reward the "fundamentally sound" plays you think it will.

Restricted AI model access = bigger security risk

@accelxr breaks it down:

If you don't get early access to frontier models, you can't build defenses before release. The second it drops, everyone's scrambling to patch their infrastructure.

This creates a dangerous window where attackers who DID get access have a massive advantage.

The irony? Trying to make AI "safer" through gatekeeping might actually make the ecosystem MORE vulnerable.

Anyone building in crypto/AI infra needs to think about this. The arms race starts at launch, not before.

Nothing like a $150B market cap sitting on $7M liquidity 💀

This is the kind of house of cards that makes you wonder who's actually buying at these levels. When the exit door is that narrow and everyone realizes it at once... well, you know how this movie ends.

Classic late-cycle behavior. Ponzinomics at their finest.

$ZEC finally rips after a decade of bagholding, just in time for AI to potentially end civilization. Poetic justice or cosmic joke? Either way, the timing is absolutely unhinged.

90% of my text today came from voice → @WisprFlow did the rest.

Only typed 10%.

This is legitimately game-changing for workflow. Voice-to-text finally doesn't suck.

Props to the Wispr Flow team — this product actually delivers.