In "The Truth About My Prop Firm" (~13:40), asked directly whether Lark copy-trades its traders' positions into a real market, Matt L. answered: "every account you receive is a demo account... you're never getting access to real funds. However, that doesn't mean that a prop firm may not actually use your data and route it to their corporate account and then copy trade via that route. So, have we done that in the past? Absolutely. Now, it's important for you to know as a trader that it's rare... Don't bank on a firm actually [doing it]... it shouldn't make a difference to you whether it's getting copy traded or not in the live market."
Read that carefully: every account is confirmed demo capital, and mirroring a trader's position into a real market, the one thing that would make a trader's profit come out of the market instead of out of Lark's own pocket, happens only "rarely," at Lark's discretion, disclosed nowhere but this one answer in a YouTube video. That means most trader P&L on this platform is a direct transfer between trader and firm: a trader's win is Lark's real cash outflow, a trader's loss is Lark's real cash inflow, not a pass-through of market profit. That's a structural conflict of interest sitting underneath every discretionary clause already documented on this page. A firm that mostly isn't hedged has a direct financial stake in finding a reason not to pay, which is exactly the latitude the "All-or-Nothing" rule and the sole-discretion Terms give it.
To be fair, the industry-standard version of this isn't "nobody hedges anything." The common structure across this industry is two entities: a sales/management brand in a reputable jurisdiction (UK, UAE, US, Canada) out front, paired with a broker entity sitting on cheaper regulation that actually executes internally. Because most challenge attempts fail statistically, the firm doesn't need to hedge them, it only needs to selectively mirror the minority of traders who prove themselves consistently profitable, so a real payout gets funded by a real market gain instead of coming straight out of the firm's own pocket. That's the version of running a mostly-unhedged book that's at least internally coherent, and it's common enough that it isn't in itself a red flag. What Matt L. describes isn't that: not a disciplined system that mirrors proven winners, but something he calls "rare" and tells traders flatly not to expect, track record or not. That's a lower bar than the industry's own usual justification for running an unhedged book, and it's the CEO's own words saying so, not an inference from marketing copy or a corporate structure we can't see. We have no evidence on what Lark's actual broker/execution setup looks like behind the scenes, and aren't going to guess at it, this finding rests entirely on what he said, not on a theory about who's on the other end.
The public Terms of Use you accept at checkout is a general platform agreement. The actual Trader Agreement, the document governing your specific funded account, payout math, and what counts as a breach, is not published anywhere. Per Lark's own help articles, it only reaches you once you've paid for and passed an evaluation: DocuSign emails it to you alongside a Veriff KYC link, and a payments processor called Deel handles the paperwork and payout processing from there. Only after that agreement is signed and KYC clears do you get login credentials for the funded account.
Reading a contract only after the money is already spent and the evaluation already passed is common practice among prop firms generally, so treat this as "worth knowing before you pay," not as something that singles Lark out. It does mean you cannot compare the actual funded-account terms against a competitor's before choosing where to put your money.
Lark prohibits "All-or-Nothing" trading, defined as a style where a trader "could breach, or come close to breaching, the drawdown limits in a single trade" through outsized position size or margin use. Their own materials cite a working guideline of roughly 20% margin utilisation, used "consistently," as the threshold that gets a trader flagged. Neither "come close to" nor "consistently" carries a number.
That ambiguity isn't theoretical: multiple trader reports describe payouts denied under this rule after an evaluation was already passed, including one citing roughly $8.7k in profit denied over a margin-utilisation call. As with any discretionary risk clause, the rule only bites at payout time, after you've already put in the work.
CEO Matt L. runs the pricing pitch on his own YouTube channel the same way in video after video: higher price buys fewer rules. In "The Prop Firm Industry Is Collapsing? Prop Firm Owner Reacts" (~11:45): "At Lark Funding, we have been in business for almost 4 years, and we have none of those rules. No consistency rules, no minimum trading days, no news restrictions, no consistent lot size rules... Yes, our challenges are more expensive." In "$1 vs $100 vs $1,000 Prop Firm Challenge" (~15:56), the pitch for the pricier tier is "flexibility... freedom... you don't have to go through all those loopholes to get paid."
Those specific claims, no consistency rule, no minimum trading days, no news restriction, check out against the published rules. What the pitch leaves out is that the FAQ covering the margin-excess flag above also carries a "recommended" per-trade risk guideline: "a maximum risk per simulated trade idea... up to 1.5%," stacked on margin-utilisation caps (50% forex/metals, 75% everything else). None of that is a hard numeric rule in the marketing sense either, it's framed as a recommendation, but the ToS's sole-discretion breach language lets Lark enforce it after the fact exactly like a rule.
That's worth flagging because the same CEO calls this exact pattern out as a problem when other firms do it. In "Drama With FTMO?" (~3:39), on FTMO's undisclosed risk cap: "there's been a lot of drama about FTMO restricting traders to 1% risk per trade. It's not on their website as like a standard rule. It's if you risk too much, then they implement that rule on you." And describing what he frames as bad industry practice generally, in the same "Industry Is Collapsing" video (~12:20): "you might still deny your payout if we decide your trading pattern looks suspicious." Lark's own All-or-Nothing clause, an undefined "come close to breaching" trigger enforced at the firm's sole discretion, is functionally the same mechanism.
Lark's homepage advertises a "Smart Reset Guarantee" for funded accounts on both evaluation tracks: "1-Step accounts get a free reset, and 3-Step accounts get a 75% discount... Trade with confidence knowing you have a safety net." That's a different mechanic from the evaluation-phase retry covered on the 1-Step page, this one applies after you've already passed. The help centre article that actually governs it states plainly: your case gets manually reviewed and "you will be emailed back regarding our decision within 72 hours," and, in the same article, "an approved Simulated Funded Account reset is not guaranteed." A homepage banner using the word "Guarantee" pointing at a policy whose own text disclaims that exact word is about as direct a contradiction as this page has found.
The eligibility criteria compound the problem. Rather than a hard numeric bar, the article leans on subjective language, "more demo trading days the better," "consistency in execution," alongside a loose reference to keeping any single loss near 1%. Run that 1% figure against the account's own drawdown ceiling (7% on the 1-Step funded account, 5% on the 3-Step): breaching from a couple of honest mistakes within that guideline isn't really possible, it would take something like 5 to 7 consecutive losses at the guideline size to burn through the whole budget, not one or two. So a trader who actually breached while following the risk guideline needed a real losing streak, not "a mistake," which is exactly what the subjective "consistency in execution" language gives Lark room to deny. And a trader who breached in only 1-2 trades almost certainly did it by exceeding the 1% guideline, an exposure violation, which is the "All-or-Nothing" territory covered above. Either way, the set of traders who both genuinely qualify and get approved is narrow by construction, dressed up as a safety net on the homepage.
The refund policy is narrow: "After a cleared payment on the purchase of one of our programs occurs, you will receive an email with the login details to access your trading platform. Once this information is emailed to you, no refund will be given." The only carve-out is an untouched account with zero trades placed, handled case-by-case through support. Separately: "Clients who improperly dispute charges or request chargebacks with their bank will be permanently banned from the Platform." Both terms are common in this industry, but the chargeback clause in particular means your only real recourse on a disputed breach is Lark's own support, not your card issuer.
Every withdrawal through Riseworks costs a flat $40, deducted regardless of size. Lark's own minimum payout is $100, so at the floor that fee alone eats 40% of what you're withdrawing. Against the firm's own published average payout of $1,225, the same $40 works out to roughly 3.3%, a fee that hits small, frequent withdrawals far harder than large ones.
We don't have access to the actual distribution of individual payout sizes, Lark's payout tracker is a live feed, not something we can audit statically, so we can't say what share of real payouts land near that $100 floor versus the $1,225 average. But the structural incentive is worth flagging on its own: this is a firm that markets itself on having no minimum trading days and no consistency rule, both of which make it easy to withdraw early and often on small gains, into a fee that's most expensive, in percentage terms, exactly when you do that.
Lark discontinued its 2-Step evaluation track in early 2026; trader reports describe funded accounts on that track (one citing roughly $150k in combined funding) closed as a result. A separate report describes a trader with roughly $400k across multiple funded accounts terminated for using bridge software to copy trades between their own accounts, something the Terms do prohibit ("copying between internal accounts" and unauthorised copy trading generally). These are individual trader accounts of events, not independently verified by SwingFish, but they line up with the broader pattern: broad discretionary clauses ("the Company reserves the right... in its sole and absolute discretion") mean program changes and enforcement calls sit entirely on Lark's side.
It would be misleading to only list the above. Lark publishes a running payout tally, on the order of half a million dollars across 400+ payouts at time of writing, with reviewers commonly reporting turnaround measured in hours rather than days. A meaningful share of Trustpilot reviews are genuinely positive about speed and support. The complaints that exist cluster specifically around the vague margin rule and program changes, not around the firm refusing to pay when the rules are clear-cut.
Lark's own site and third-party trackers currently disagree with each other on fee and drawdown numbers for the same account tier (for example, a $10k 1-Step evaluation is listed both as a $200 fee with 7% drawdown and as a $125 fee with 6% drawdown, depending on the source). Rather than publish a number we can't stand behind, check the live checkout at larkfunding.com directly before buying.
