Invest1 publisher3 min readPublished Updated
ABFinance had the US licences and folded before it ever launched
The startup from ex-Bybit co-CEO Helen Liu says it will not launch, months after unveiling. In a year with about 95 crypto project closures, the constraint has moved from permission to distribution.
The Investor · Invest desk
Drafted by a language model from the sources cited here and checked against its claim ledger before publication. How we use AISend a correction

What happened
- ABFinance said on August 14, in a post on its X account, that its planned launch will not move forward and that it is winding down in an orderly manner.
- ABFinance, a US finance startup, was founded by former Bybit co-CEO Helen Liu and unveiled in mid-March, days after Bybit confirmed her exit.
- Liu was building the platform in the United States with full US licensing in place from day one.
- The planned product was a single platform handling deposits, trading, earning and payments while linking fiat currency to crypto.
- No specific reason was given for the shutdown, and the company did not say what would happen to its education programme.
Compiled by The InvestorSomething wrong?How this is made
Why it matters
ABFinance, the US finance startup founded by former Bybit co-CEO Helen Liu, said on August 14 in a post on its X account that its planned launch will not move forward and that it is winding down in an orderly manner [1]. The company was unveiled in mid-March and, per Cryptopolitan's account, was being built in the United States with full US licensing in place from day one [2][3].
That last detail is the whole story. The plan was one platform covering deposits, trading, earning and payments, bridging fiat and crypto [4], which is precisely the product that spent the last cycle being described as blocked by regulators. ABFinance was not blocked. It had the permissions and stopped anyway, and it gave no specific reason for doing so [5], which means anyone attributing this to rulemaking is guessing against the only fact on the table [5].
Licensing is a gate you pay for once. Customer acquisition is a bill that arrives every month, and a deposits-plus-trading-plus-payments product has to win retail users away from firms that already have them. Cryptopolitan attributes this year's closures to a prolonged bear market and to businesses merging into fewer large platforms [10]. Consolidation is another way of saying that distribution has already been allocated. A licence gets you the right to compete for it; it does not lower the price.
The surrounding tally is not kind. By late July, Cryptopolitan counted 17 major crypto ventures closed in 2026 and roughly 95 projects in total [9], so the headline failures are about 18 percent of the count and the rest are quieter [2]. Bitmart said in July it would close worldwide operations after nine years [11], BitMEX ended an 11-year run around the same time [12], and Syndicate Labs, Everclear and ZERO Network closed within hours of each other in May [14]. Bitwise cut about 14 percent of staff in August after its main index fund's net assets fell 48.27 percent to $532.8 million in the first half [13], which implies the fund started the period near $1.03 billion [3]. None of those were licensing problems either.
Two dating wrinkles are worth noting because the source contains both. Cryptopolitan describes the wind-down as coming about four months after the unveiling [8], but mid-March to August 14 is closer to five [1]. It also says ABFinance was unveiled days after Bybit confirmed Liu's exit while dating her step-down to April 30, 2026, after five years at the exchange [2][15], leaving roughly six weeks between the announcement and the departure date [4]. Treat the four-month framing as approximate.
The unresolved item is ABF Learn, described as the company's only shipped output: an education programme for students and early-career people covering how banking, AI and digital assets fit together [6]. The wind-down statement thanked community, partners and the people who built alongside the company [7] but said nothing about the programme's fate [5].
Watch whether the licensed entity and its permissions are sold rather than dissolved, since that is where residual value sits in a pre-launch bank. Watch whether ABF Learn is transferred or simply switched off [5]. And watch the ratio of major to total closures [2]: if the 17 keeps climbing against a flat 95, the failures are moving up the size curve, which is a different problem from a long tail giving up.