Invest3 distinct publishers2 min readPublished
Brent Kovar's Profit Connect ran on a supercomputer that did not exist and on guaranteed returns paid out of new deposits. Measured illicit crypto flows suggest the type is not rare.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
A fixed 15% to 30% a year is a payment schedule before it is a marketing claim. On $24 million of principal, that promise carries between $3.6 million and $7.2 million of annual obligation [20], and the court-appointed receiver who took over Profit Connect in 2021 found the claimed mining had produced only negligible results [14]. Prosecutors said payouts were funded by money from newer investors and presented as proceeds of crypto activity, while investor cash also covered company expenses, employee gifts and a house for Kovar [8]. FBI special agent Christopher Delzotto said victims "thought they were engaged in revolutionary technological advancement" when the operation was "merely a deception" [19].
The diligence question in this case was never technical. When the SEC froze the company's assets in July 2021 [9], it said more than 90% of what Profit Connect had raised came from investors rather than from trading [10]. An allocator needs no opinion on whether software can mine profitably to ask what share of distributions is funded by customer revenue against new subscriptions. That ratio was on the public record in 2021, and it is the ratio to ask of any operator selling a guaranteed yield now.
The timeline carries the other useful number. The indictment came on 14 February 2025, about three years and seven months after the freeze, and sentencing is set for 30 November 2026, roughly five years and four months after it [12][13][23]. In that window the measured harm doubled. The SEC described over $12 million from more than 277 retail investors [10]; the criminal case established at least $24 million from at least 400 [7], twice the money and 44% more victims [22].
Scale sets the last limit on what one verdict can do. TRM Labs estimates investment into the illicit crypto market reached a record $158 billion in 2025, up nearly 145% on the year [15], and Chainalysis counts at least $14 billion received on-chain by scams in 2025, a figure it expects to pass $17 billion as more addresses are identified [16]. The two measure different things, but at the Chainalysis rate, Profit Connect's entire four-year take is about fifteen hours of scam inflows, or 0.17% of a single year [21]. INTERPOL flagged the "industrialization of fraud" in March 2026 [17]. A nine-day trial and five years of process, ending in one conviction [1], does not bend a base rate of that size. The screening stays with whoever writes the cheque, and the SEC's allegation that Profit Connect pushed people toward retirement savings and home equity [11] indicates who tends to write it without asking.
Ranked by verification strength, evidence, and original report placement.
A federal jury in Nevada found Las Vegas businessman Brent C. Kovar guilty on 11 counts of wire fraud, two counts of mail fraud and two counts of money laundering after a nine-day trial.
Kovar was convicted on August 24; the US Justice Department announced the verdict on a Monday.
Kovar owned Profit Connect from late 2017 to July 2021, and the company purportedly used artificial intelligence software on a supercomputer to mine crypto and verify crypto transactions.
Profit Connect presented itself as an artificial-intelligence company mining cryptocurrency on a supercomputer; that technology did not exist.
Kovar falsely presented Profit Connect as profitable and promised fixed annual returns ranging from 15% to 30% along with a 100% money-back guarantee.
Kovar claimed investor deposits were protected by the Federal Deposit Insurance Corporation and that Profit Connect held hundreds of millions of dollars in crypto reserves; both claims were false.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Adjudicated court record plus a documented civil prior
The core facts rest on a jury verdict after a nine-day trial, an itemized 15-count conviction, trial evidence on payment flows and personal spending, a 2021 court-appointed receiver's finding of negligible mining, and an SEC emergency freeze with specific fundraising figures. Three independent publishers report the same load-bearing numbers. The residual uncertainty is narrow: the FDIC-insurance representation is characterized more firmly by one publisher than another, and the wider market statistics are single-sourced secondary reporting of vendor estimates rather than primary documents.
Real victim uptake, zero real technology deployment
Adoption here has two distinct readings and both are measured in the sources. The fraudulent offering itself achieved meaningful uptake: at least 400 investors and more than $24 million over roughly three and a half years, roughly double the money the SEC had counted at the July 2021 freeze. The claimed technology achieved none - the receiver found negligible mining output and prosecutors showed virtually no legitimate revenue. The broader pattern is quantified only through vendor aggregates, and against those the case is small: about 0.17% of Chainalysis' 2025 on-chain scam receipts.
Case facts grounded; market-scale framing extrapolated
The scheme's own claims were fully overstated - a supercomputer that did not exist, reserves that were not there, insurance that was never in force - but that gap is now adjudicated rather than asserted, so it does not inflate this cluster. The modest positive score comes from the framing layer: one publisher generalizes from a $24 million case to a $158 billion illicit market and INTERPOL's 'industrialization of fraud' using single-sourced vendor and multilateral estimates, and the 280-year statutory maximum is foregrounded in headlines even though every source concedes the judge sets the actual term under guidelines. The case-level reporting itself is close to aligned with the record.
Enforcement publicity and vendor-supplied market statistics
Attributed statements in the cluster come from parties with visible interests: DOJ, FBI, FDIC OIG and the US Attorney's Office are quoted framing the conviction as a deterrence win, and the scale statistics come from TRM Labs and Chainalysis, commercial blockchain-analytics firms whose market grows with the perceived size of illicit flows - a conflict none of the publishers disclose. All three outlets are crypto-native trade press, and the widest-framing piece explicitly ties enforcement success to mainstream adoption and institutional confidence, an angle aligned with their readership. Offsetting this, the central facts are court-adjudicated rather than sourced from interested parties.
High on the case, moderate on the surrounding market claims
Three independent publishers converge on the verdict, counts, dollar and victim totals, procedural dates and the nonexistence of the claimed technology, all anchored in an adjudicated record - that portion is high confidence. Confidence drops for the contextual layer, which depends on one publisher's secondary reporting of vendor and multilateral figures, and for details such as the exact characterization of the FDIC representation and the ultimate recovery for victims, which the sources do not resolve.
invest
DOJ Flew a $165M Ponzi Case Home From Fiji, Which Resets the Affiliate Math1 distinct publisher
invest
Washington licenses private hacking, and hands the contractor the liability1 distinct publisher
invest
Singapore's exchange raids stop looking like incident response2 distinct publishers
invest
Washington deputises private cyber firms, and hands their customers a liability question1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
cointelegraph.com
1 article · August 25, 2026
crowdfundinsider.com
1 article · August 25, 2026
cryptopolitan.com
1 article · August 25, 2026