NexFlow › Is Uphold safe

Is Uphold safe? The transparent ledger that promoted a loan book it didn't understand

Uphold's safety story inverts the usual order: the exchange layer — reserves published live, never breached on record — is the strong part, and the failure lives in what Uphold *sold*. Between 2019 and 2020 it pushed CredEarn to its users as 'safe' and 'insured'; the loans behind that yield went to a Chinese micro-lender writing uncollateralized 14-day loans to video gamers. When Cred collapsed, ~6,000 Uphold customers were inside. Years later the New York AG extracted $5M for the marketing itself. Whether Uphold is safe depends — again — on which product you used.

Updated September 28, 2026 · By the NexFlow editorial desk · Sources linked throughout; vendor claims labeled.

What Uphold is

Uphold is a multi-asset platform founded 2014 (as Bitreserve) — crypto, fiat, and metals on one ledger, roughly ten million users, headquartered across US and UK entities. Its signature practice is the thing most of this corpus asks for and rarely gets: a real-time reserve status page showing assets against obligations, a transparency posture it pioneered years before 'proof of reserves' became an industry checkbox.

Custody at the exchange layer is the standard custodial model — Uphold holds keys for platform balances — and the documented record there is clean: no exchange-level breach draining customer funds is on file across a decade-plus. The instructive failure, which is why this page exists, sits one shelf over: the products Uphold chose to promote to those same customers.

CredEarn: the yield product that wasn't what the ads said

In 2018 Uphold partnered with a lender called Cred and, from January 2019 through October 2020, promoted CredEarn inside its app — ads, pop-ups, a featured spot on its integrations page — as a way for customers to earn significant interest on their crypto. Per the New York Attorney General's later findings, Uphold marketed it as a 'safe,' 'reliable,' 'insured' savings product; roughly 6,000 Uphold customers put approximately $50 million of crypto into it.

What the yield actually was: Cred routed the crypto to MoKredit, a Chinese micro-lender — part-owned by Cred's own co-founder — writing uncollateralized 14-day micro-loans at rates often above 35% to young Chinese video gamers with low incomes and no credit history. Uphold understood this by its own later admission; it did not tell customers that is what 'comprehensive insurance' and 'reputable companies' meant in practice — and per the AG, no insurance protecting retail investors' digital-asset losses existed at all.

MoKredit stopped repaying principal in March 2020. Cred filed Chapter 11 in November 2020; the customer crypto inside CredEarn went into a multi-year liquidation. The Cred Liquidation Trust later sued Uphold for at least $783 million, alleging it jointly designed the product (Cred's founder sat on Uphold's board; the suit claims it was nearly named 'UpholdEarn' before a rename to dodge regulatory risk) — a claim the bankruptcy court dismissed in April 2023 and the district court affirmed on appeal. Uphold was not legally made the owner of Cred's losses — the customers absorbed them through the liquidation instead.

What the regulator actually settled

The legal dismissal matters and so does what came after: the New York Attorney General's investigation concluded Uphold's *promotion* — not the loans, the marketing — misled investors, and extracted a settlement of more than $5 million: restitution to harmed users plus a requirement that Uphold change its third-party-product policies. Five million dollars is more than five times the fees Uphold earned on the product; the AG's framing was that you cannot wrap a stranger's loan book in your brand's credibility and disclaim the wrapping when it fails.

That is the Gemini Earn shape a year earlier and a tier lower — and sharper in one respect: a regulator didn't just document the counterparty collapse — it put a price on the advertising language itself, which is the precedent every venue marketing desk in this industry should be reading.

The exchange layer vs. the product shelf

Hold the two records separately. On custody: the reserve dashboard is real, the segregation is real, and there is no exchange breach to report — Uphold's core is arguably better-evidenced than bigger names in this family. On product judgment: the one yield product it chose to carry became the corpus's cleanest 'the venue was fine, the promoted product wasn't' case — with a regulatory finding that the sales language was the offense.

The practical reading a user should take: an exchange's live-reserve page tells you the platform's own books balance at that refresh. It says nothing about a third-party product its marketing team decides to feature, and CredEarn is the documented proof that those are separate questions even on the most transparent ledger in the industry.

Where Uphold stands

In the mid-tier custodian cluster Uphold is the transparency pioneer with a promotion-record asterisk: strongest-in-tier on reserve disclosure, unbreached at the custody boundary, and carrying a $5M-plus regulatory settlement whose subject was the gap between what it sold and what it verified. For holding assets on-venue, the record supports the tier's top marks; for any yield or third-party product the platform features, the CredEarn file is the caution priced in.

Frequently asked questions