Arthur Hayes has a message for the token launch playbook the industry has run for years: it’s broken.
The Flop Labs CEO said high-FDV, low-float tokens no longer pump on Binance. He also argued that crypto has abused tokenomics through venture capital presales that leave retail investors holding the bag.
What Hayes is actually complaining about
Some quick vocabulary first. FDV, or fully diluted valuation, is what a token would be worth if every token that will ever exist were already circulating.
Float is the slice that’s actually tradable today. A high-FDV, low-float launch means a big headline valuation sitting on a small pile of tokens people can actually buy.
In token terms, those new tables are unlocks. Early backers, often VCs who bought in at presale prices, gradually receive tokens they can sell into the market.
According to the research findings on his remarks, Hayes argued this structure incentivizes those early investors to take profits. The people on the other side of those trades are retail buyers in the secondary market.
The model has dominated the industry for more than a decade, going back to the ICO boom, per the same findings. Hayes’ point is that the market has stopped falling for it. A Binance listing used to be a launchpad. For these tokens, he says, it no longer is.
The Flop Labs alternative
Hayes laid out his case during a keynote at TOKEN2049 Singapore on October 7, 2026, where he presented Flop Labs’ approach. The company calls it a “100% fair launch.”
The concept strips out the parts of the standard playbook he criticized. There are no presales and no VC allocations.
Instead, tokens get distributed through airdrops tied to network participation. Users earn FLOP by doing meaningful work on the testnet rather than by writing an early check.
The numbers are specific. Approximately 24.3% of the projected year-10 supply, about 4.4 billion out of 18.1 billion tokens, is earmarked for a genesis airdrop.
Timing is also mapped out, though still framed as a plan. The airdrop is scheduled to start in late October or early November 2026 and run for around 90 days.
After that, Flop Labs plans a mainnet genesis in Q1 2027. FLOP tokens are not tradable yet and won’t be until after the airdrop.
Funding is the other notable detail. Flop Labs is financed through Hayes’ personal funds, which is consistent with a pitch built on keeping VCs off the cap table.
Why AI agents are part of the argument
Hayes argued that agents operating within the network will need utility-based tokens linked to computational tasks. In his view, existing fiat currencies and stablecoins aren’t the right fit for that job.
The pitch, put simply: if software agents are buying computing power from each other, they need a currency that converts directly into that compute. A dollar-pegged stablecoin measures value in dollars. A compute-linked token measures it in work done.
The research findings frame this as aligning token issuance with real-world computational demand, particularly as AI applications need a decentralized economic layer.
Hayes returns, with a warning
Flop Labs marks Hayes’ return to building in crypto after stepping back from BitMEX.
He also issued a practical caution. Hayes urged people to watch out for scams promoting early access to FLOP tokens.
Since FLOP isn’t trading and won’t be until after the airdrop, any offer to sell it early is, at best, wishful thinking.
What this means for traders and token launches
For retail traders, Hayes is saying out loud what charts have been suggesting. If high-FDV, low-float launches no longer produce the listing-day pop, the old strategy of buying the Binance listing and waiting stops working.
Flop Labs is effectively a live test of the alternative. Allocating roughly a quarter of year-10 supply to participants through an airdrop is a deliberate bet that community distribution beats insider allocation.
The concrete markers to watch are the airdrop start in late October or early November 2026, roughly 90 days of testnet participation, and the planned Q1 2027 mainnet genesis.
Disclosure: This article was edited by John Chen. For more information on how we create and review content, see our Editorial Policy.