Blog – Full Posts

The Bayou Hedge Fund fraud turned credibility into a weapon

Earmark Team · September 14, 2026 ·

On June 9, 2008, a white GMC Envoy sat abandoned near the Bear Mountain Bridge, about 40 miles north of New York City. The keys and a bottle of pills were inside. Traced through the dust and pollen on the hood were three words: Suicide is painless.

The vehicle belonged to Sam Israel III, who was supposed to report to federal prison that morning to begin a 20-year sentence. Authorities searched the Hudson River but found no body. Israel had staged the scene and fled in an RV.

It was a fitting final act for a man who had spent years telling investors what to believe. In Episode 118 of Oh My Fraud, host Caleb Newquist explains how Sam and his partners built Bayou Hedge Fund Group on fake performance, false audits, and borrowed credibility. Investors ultimately lost about $300 million.

A famous name opened doors that performance could not

Sam inherited credibility before he earned it. His grandfather built a coffee importing company into ACLI International, a commodity trading business later acquired by Donaldson, Lufkin & Jenrette for $42 million. The Israel family name carried weight on Wall Street.

Sam’s résumé didn’t deserve the same confidence. He claimed he’d served as head trader at Omega Advisors and managed more than $400 million. When someone called Omega founder Leon Cooperman, the story fell apart. Sam had worked there for about 18 months and had no trading discretion.

In 1996, Sam launched Bayou with James Marquez and Daniel Marino, a CPA. They promoted a proprietary system called “forward propagation,” which supposedly found patterns other traders missed. It sounded impressive, but investors couldn’t test it—and it didn’t work.

That failure set the stage for a much larger deception.

Bayou’s independent auditor was anything but independent

By the final trading day of 1998, Bayou had suffered heavy losses. Sam, James, and Daniel decided to report profits they hadn’t earned, attract more capital, and trade their way out of trouble.

To support the lie, they used Richmond Fairfield Associates as Bayou’s auditor. The firm sounded established, but Daniel controlled it while also serving as Bayou’s CFO. He helped prepare Bayou’s financial information and then issued supposedly independent opinions on it.

“This isn’t a case where the auditor failed to catch the fraud,” Caleb says. “The auditor was the fraud.”

The fake audit helped Bayou report a 17% return for 1998, including a 3% gain in December. Bayou charged investors 20% of its reported profits and routed most trades through Bayou Securities, an affiliated brokerage controlled by Sam. That brokerage collected nearly $3.3 million in commissions from Bayou fund activity between 1997 and 2000.

The SEC later said Bayou never produced a genuine year-end profit. Still, the paperwork made the fantasy look real.

A flood of information created the illusion of transparency

Sam sent investors detailed letters about markets and performance. The communication felt reassuring, but, as Caleb notes, “Receiving a lot of information is, of course, not the same thing as receiving accurate information.”

Other credibility signals piled up:

Marketers earned as much as 3% of the assets they brought in, with payments continuing while clients remained invested
Professional advisers recommended Bayou, making it appear thoroughly vetted
Bayou identified Grant Thornton as its auditor in 2002, although the firm later said it hadn’t worked for Bayou since the late 1990s
Bayou routed trades through a brokerage controlled by its founder

Some investors asked questions. Tremont Capital Management withdrew after Bayou couldn’t explain why related funds reported different returns. Most investors stayed, however, and Bayou’s fabricated performance kept attracting money.

Soon, fake profits began producing very real compensation.

A $92 million discrepancy exposed the scale of the fiction

In 2003, Bayou launched four new funds and attracted more than $125 million. That year’s results show how far the reported numbers had drifted from reality:

Actual trading result: a $49 million loss
Reported result: a $43 million profit
Difference: $92 million

Sam and Daniel collected incentive fees based on those invented profits. By spring 2004, Bayou claimed more than $350 million under management even though the SEC said it had stopped almost all securities trading.

With roughly $150 million left, Sam moved nearly all of it into supposed “prime bank trading programs.” These secret, low-risk, high-return markets didn’t exist. Arizona authorities eventually froze about $101 million, creating an outside paper trail Bayou could no longer control.

Once outsiders controlled the records, the fraud began to collapse.

A confession ended the fund, but not the spectacle

In August 2005, investor Eric Dillon entered Bayou’s empty Stamford office and found Daniel’s six-page letter. It began, “This is my suicide note and confession.” Daniel wrote that he, Sam, and James had defrauded investors since about 1998.

Sam and Daniel pleaded guilty, and each received a 20-year sentence plus a $300 million restitution order. James received 51 months and was ordered to pay more than $6 million.

Then Sam staged the bridge scene. After about three weeks hiding in the Northeast, he surrendered at a police station in Southwick, Massachusetts, arriving on a motorized scooter. The escape added two years to his sentence.

Authorities returned more than $150 million to victims in 2008 and another $31.8 million five years later. Bayou’s bankruptcy estate also pursued investors who had withdrawn money near the end of the scheme, seeking more than $135 million in redemption payments.

The fallout eventually reached the advisers who’d helped make Bayou look credible.

Verification separates evidence from theater

In 2009, the SEC charged the Hennessee Group and principal Charles Gradante with failing to perform parts of the due diligence they’d advertised, including properly investigating Richmond Fairfield. They settled for more than $814,000 without admitting or denying the findings.

The Bayou story offers some lessons for accounting professionals:

Confirm résumés and service providers directly
Test auditor independence rather than trusting a professional-sounding name
Investigate related-party brokerages and compensation arrangements
Reconcile performance reports with bank, custodial, brokerage, and trading records the manager doesn’t control
Treat unusually smooth returns and excessive client-generated paperwork as reasons for more scrutiny

Bayou survived because each link in its credibility chain assumed someone else had already checked. Transparency isn’t measured by how much paper a client provides. It depends on whether that paper leads to independent evidence.

For the complete story, including the full absurdity of Sam’s attempted escape, listen to the Oh My Fraud episode.

Your Firm Feels Broken When the Business Model Doesn’t Match Your Values

Earmark Team · September 14, 2026 ·

Between 2017 and 2019, Sandra Koch was searching for a way to fix what she called her “broken accounting firm.” She tested popular industry models, including selling tax plans. Instead of finding clarity, she felt confused and stressed. The approaches required her to become someone she wasn’t.

That experience raised an important question: What if recurring friction isn’t a sign that you need more discipline, another app, or a better implementation plan? What if the model simply doesn’t fit your values?

In a recent episode of Who’s Really the BOSS?, Sandra explains how she shaped Aurora Consulting Group around her beliefs about clients, employees, and work. Founded in San Diego in 2011, the firm is now based in Visalia, California. Sandra lives in Baja California Sur, Mexico, and leads a fully remote, six-person team spread across Mexico, California, and Indiana.

Aurora earns revenue in the low $700,000 range. It primarily serves service businesses with $1 million to $5 million in revenue, providing accounting, tax, advisory, and operations-management support.

Stop forcing someone else’s model onto your firm

For years, Sandra operated with a small team, but without the professional community and support system she later realized she needed. She looked outside the firm for answers and tried several approaches that didn’t work for her.

“If I try to twist myself into something I’m not, that’s inauthentic,” she says. “It’s going to fail.”

Her point isn’t that tax planning or any other service model is wrong. The problem comes when a model clashes with what you believe about client service, leadership, or work. Even strong execution can’t make a poor fit feel natural.

When your firm keeps “grinding gears,” consider asking:

What do we believe excellent client service requires?
What kind of workplace do we want to create?
Which services truly help our clients?
Do poor execution or poor alignment cause this frustration?

Once your values are clear, technology and operating decisions become easier to evaluate.

Let technology support the work

Aurora uses a practical, cloud-based technology stack. Karbon holds client work and email. Clients use QuickBooks Online, while the firm uses ProConnect Tax, Ignition, Ramp, and either Gusto or Rippling. The team also checks Google Workspace before purchasing another tool. Google Forms, for example, often meets its needs without an added subscription.

Sandra adopted cloud software early. After H&R Block purchased RedGear Technologies in January 2012, she had about two weeks to replace her tax software. Intuit was the only company that could import her client data within that window, so she moved to ProConnect. She also began testing QuickBooks Online before it included bank reconciliation because she believed Intuit would continue improving it.

The firm takes an equally practical approach to AI:

Blue J supports tax and accounting research and helps explain technical topics to clients
ChatGPT and Claude support creative, subjective, and organizational work. For objective claims, Sandra says, “It’s our job to prove it.”
EasyLlama provides security, AI, empathy, and customer service training in one place

Technology supports remote work, but a healthy workplace requires trust and attention to people.

Build flexibility around real lives

Every Aurora role has a job description, but Sandra recognizes every employee is different. When one employee’s abilities and strengths didn’t match the role or Sandra’s expectations, the mismatch stressed the team.

Instead of pushing the employee to perform within the same structure, Aurora stripped down and rebuilt the role around the person’s strengths. “We couldn’t see it because of the chaos the mismatch caused,” Sandra explains.

That same care shapes the firm’s schedule. Everyone must attend the Tuesday team meeting at 10 a.m. Beyond that (and California wage-and-hour requirements), Aurora doesn’t require employees to work set days or hours. Employees block unavailable time on their calendars, communicate their schedules, and remain available for necessary meetings.

That flexibility allowed one employee to work half days for two weeks so she could attend her child’s playoff games. Sandra reasons the employee won’t regret those half days years from now, but she would regret missing the games.

However, flexibility only works when clear service standards protect clients.

Turn responsiveness into a shared responsibility

Aurora uses Grasshopper for its phone system. Calls to the general line ring across the company, so any employee can answer. Everyone also monitors the shared client text line. A written policy says that if a message remains unanswered for an hour, someone must alert the intended recipient.

Karbon helps the firm track email response times, and every employee includes a booking link in their email signature. Each client also works with three contacts: a client service manager, a controller, and a CFO. If one person is unavailable, the relationship doesn’t stall.

Podcast host Rachel Dillon highlighted a lesson firm leaders often overlook: Employees can’t follow a rule that exists only in the owner’s head. If the same issue keeps causing frustration, ask whether you’ve documented and communicated the expectation.

But responsiveness is about more than speed. When Aurora learned that a client had lost a family member, the team shared that context so everyone would approach future conversations with care. As Sandra’s story shows, people remember when somebody notices.

Give every team meeting a clear purpose

Aurora’s Tuesday meeting begins with a team member reading the firm’s mission and values. Employees then share work highs and lows, discuss what they need help with, and celebrate progress. The meeting also includes a book discussion, announcements, personal milestones, and “happies and crappies.”

Client problems are intentionally absent. Aurora once discussed them during staff meetings, but the meetings became too long and left employees discouraged. Client issues now go to separate meetings. The weekly team meeting’s purpose is to be a pep rally that builds connection.

That reflects Sandra’s larger message, which is every system, from meeting agendas to response policies, should support the way you want to work.

Build from conviction, not convention

Sandra’s experience offers five lessons:

Define your values before choosing services, systems, or management practices
Treat recurring friction as a warning that something may not fit
Support flexibility with clear schedules and written standards
Shape roles around people’s strengths whenever possible
Keep technology purposeful and verify objective AI-generated information

“When you match your firm to your value system,” Sandra says, “everything’s just smooth.”

Your firm doesn’t need to look like everyone else’s. It needs to serve clients well while remaining authentic and sustainable for you and your team.

Listen to the full episode to hear Sandra Koch’s full story and learn how Aurora puts these principles into practice.

Rachel and Marcus Dillon, CPA, own a national, remote client accounting and advisory services firm, Dillon Business Advisors, with a team of 28 professionals. Their latest organization, Collective by DBA, supports and guides accounting firm owners and leaders with Streamlined OS, an operating system for accounting firms, mastermind groups, and one-on-one advisory.

The Case for Building Your Accounting Career Around the Person You Are

Earmark Team · September 10, 2026 ·

Halfway through the second day of an accounting conference, Jean Zick often feels her brain slow down. The lights, conversations, and crowded schedule drain her energy. Her type-A instinct tells her to attend every session because she paid to be there. But experience tells her to take a break.

That push and pull is at the heart of Episode 35 of She Counts. Nancy McClelland and Melissa Miller Furgeson, sitting in for Questian Telka, share two conversations recorded at the Theater of Public Speaking Curtain Call retreat. Jean explains how she learned to work with her introversion. Terri Warren describes how a layoff and two strokes forced her to redefine success.

Their circumstances are different, but their message is the same. Understanding yourself gives you more choices. You can stop pushing against who you are and start building a career that fits.

Naming what drains you turns judgment into useful information

Jean was sensitive to sound as a child. She was also shy, timid, and quick to cry. Years later, after staying home while her children were young, she returned to the business world and began her career as a CFO. Networking came with the job, but large conferences left her depleted.

“This is just really draining me of energy,” she remembers thinking. “I need to do something about it.”

Jean learned that introverts and extroverts respond differently to stimulation. An extrovert may enter a bright, crowded room and think, Bring it on. An introvert may wonder, Where can I hide? Neither response is a defect.

“We are all on one big spectrum,” Jean says. “There’s nothing wrong with you. We’re just wired differently.”

Her process includes:

Identify the situations that overwhelm or drain you
Accept your response instead of asking what is wrong with you
Recognize that response as a signal
Decide what would help you participate without exhausting yourself

Nancy describes herself as an extreme extrovert and recognized the same pattern from the other side. She’d judged her own “spazzy extroversion” as a failure. Professional expectations can make anyone feel as though they show up incorrectly.

Once we stop making that judgment, we can move from shame to strategy.

Plan for participation instead of relying on endurance

Jean doesn’t try to turn herself into someone who loves entering a room full of strangers. Instead, she plans for the way her energy works.

Her conference strategies include:

Find a wing person. Ask someone comfortable meeting people to make introductions. Nancy and Sharrin Fuller have even invited conference attendees to find them when they need that support.
Attend first-timer events. At Scaling New Heights, volunteers at the first-timers breakfast help newcomers make an initial connection and continue checking on them during the conference.
Schedule breaks in advance. Don’t wait until you’re completely depleted.
Reset your body and mind. Jean may step outside, walk, drink water, and eat a snack before returning.

“In the early conferences, I felt like I was hiding,” Jean says. “When I do that now, I know I’m using it to re-energize.”

The action may look the same, but its meaning has changed. Taking a break is a deliberate way to remain engaged.

Jean also encourages introverts to recognize their strengths. They may be empathetic listeners, comfortable with silence, thoughtful questioners, and builders of deep one-on-one relationships. Nancy may briefly meet 50 people at an event, while her introverted husband ends the evening knowing what is really happening in a few friends’ lives.

Those strengths can also shape how firms build stronger teams.

Smaller spaces can create stronger working relationships

When Jean joined a large networking organization, she didn’t try to meet everyone. She joined a committee and gradually got to know eight or 10 people.

She uses the same approach in her fully virtual firm, whose employees work across 12 or 13 states. Slack’s Donut app pairs two colleagues for a 30-minute monthly conversation with one rule: Don’t talk about work. Her firm also holds a monthly team meeting, smaller client-service meetings, and check-in huddles every other week.

During large virtual meetings, employees begin with a three-to-five-minute icebreaker in breakout rooms of about three people. That smaller setting helps them warm up before returning to the full group.

For firm leaders, the lesson is not to assume one social format works for everyone. Use pairs, committees, mentorship programs, and small groups to make connection easier. When colleagues know one another, collaboration becomes more natural.

Jean learned to design work around how she’s wired. Terri’s challenge was learning to work with the person she had become.

A change in capacity calls for a new definition of success

Being laid off from public accounting “totally shattered everything I knew about myself,” Terri says. She built her identity around the idea of “Drive, drive, drive. Push, push, push. Work, work, work.”

With help from her community, she rebuilt her confidence and created success within her own walls. Her daughter captured that confidence in a bright pink cape covered with sparkles, crystals, and pom-poms.

Then Terri had a stroke and spent 12 months in rehabilitation. About a year-and-a-half later, she had a second stroke. Still determined to prove that the strokes wouldn’t define her, she worked 10 times harder for four months.

Eventually, she hit a wall. Work the old Terri could have done easily now drained her. Refusing to accept her deficits had become part of the problem.

“I was in the way,” she says.

Terri created a new scoreboard. She no longer needed “layers and layers and layers of work and clients” to prove she was successful. Instead, she could use her expertise to train and coach others. When her body told her to stop, she could put away the computer for the day.

Her cape became reversible. One side was bright pink, but the other used darker greens, blues, and purples with smaller sparkles. Terri was still stubborn, valuable, and fully herself. Her new definition of success simply looked different.

Build a career for the person you are today

Terri closes with two reminders: “Be graceful with yourself,” and use the GROW model (goal, reality, options, and will) to examine your needs, capacity, and choices.

Together, Jean and Terri offer four actions:

Notice what drains or overwhelms you
Stop treating your response as a personal failure
Put support and boundaries in place before exhaustion arrives
Revisit success when your life or capacity changes

You don’t owe the profession a version of yourself that no longer fits. Listen to the full episode of She Counts, then ask: What are you still asking of yourself that no longer fits who you are?

Nvidia’s AI Funding Deal Has “Shades of Enron,” Even If It Follows the Rules

Earmark Team · September 9, 2026 ·

The most unsettling part of a proposed $500 billion AI data-center fund isn’t that investor Michael Burry says it has “shades of Enron.” It is that, as Blake Oliver explains, nobody appears to be breaking a rule.

“There’s no fraud happening here,” Blake says in Episode 501 of The Accounting Podcast. “This is all happening in plain sight.”

That tension runs through Blake and David Leary’s discussion. AI makes financing structures more complex while helping firms complete audits and other accounting work faster. Yet many of the standards governing that work were written for a different era.

Nvidia’s financing shows how risk can grow within the rules

The proposal Burry criticized brings together Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to finance AI data centers. The plan is to use special-purpose vehicles to own the centers, buy Nvidia chips, and lease computing power to companies such as OpenAI and Anthropic. The debt would be backed by the computing assets, with Nvidia reportedly guaranteeing about 25%.

Why would Nvidia support separate entities instead of building the centers itself? Blake explains that selling chips to those entities would let Nvidia record revenue upfront. If Nvidia built and operated the centers, it would absorb the construction costs and recognize revenue later when it sold computing services.

The proposal adds another layer to the circular financing problem discussed in Episode 488. Money moves in a circle, and both sides report revenue.

Burry called the proposal an effort to use “unnatural credits to prolong momentum late in the bull phase.” “Maybe the problem is that GAAP allows this,” Blake notes.

Depreciation adds to his concern. AI chips are only useful for two or three years, but some companies use estimated lives of five or six years. Longer useful lives mean less annual depreciation and higher reported profit. Because large technology companies carry heavy weight in the S&P 500, a sharp correction could hurt ordinary investors holding index funds.

That same divide between reported results and underlying quality appears in audit.

Faster audits don’t automatically mean better audits

EY says AI improved audit speed or throughput by roughly 125% to 150%, while clients haven’t demanded lower fees. The firm also reported a 5% PCAOB deficiency rate, down from 28% the prior year, and pointed to its billion-dollar investment in people and technology.

David is skeptical that technology alone explains the improvement. Other large firms also posted better inspection results. He suggests the PCAOB’s changing focus on firmwide quality-control systems may affect the numbers.

Blake offers another theory. PCAOB inspections often focus on whether auditors followed required procedures, obtained approvals, and completed documentation. AI is well suited to checking those boxes. But it can also create work that looks “solid and sophisticated” while still being wrong. Complete documentation isn’t the same as sound professional judgment.

The productivity gains could still disrupt the market. Big Four firms may keep the savings as higher margins, but Blake argues that regional and smaller firms could eventually use the same tools to provide comparable services at lower prices.

Before that can happen safely, however, audit rules must catch up.

Audit standards weren’t built for AI agents

In a Gartner poll of 743 audit professionals, 93% reported using AI in some form. Yet only 30% used it for audit testing, 12% used it for quality reviews, and 38% of audit leaders had an AI strategy.

Hofstra University accounting professor and CPA Jack Castonguay argues that AI is audit’s biggest disruption since the corporate failures that led to the PCAOB’s creation. He says applying existing standards to a “fundamentally new operating model” won’t be enough.

The unanswered questions include:

Evidence reliability. What happens if AI invents evidence or changes data it believes is wrong?
Agent supervision. Who is responsible when auditors fail to review AI agents that gather and analyze evidence?
Independence. Could an AI-enabled accounting system and an audit platform trained on the same data reinforce the same errors?

AI can test every transaction instead of a sample. That is a major advance, but current standards don’t explain how much human review is needed when a machine examines the full population. Castonguay wants standards for acceptable use, oversight, evidence, supervision, and independence.

The mismatch is also visible in financial reporting.

Reporting and assurance are moving on different clocks

The SEC’s proposal to move public companies from quarterly to semiannual reporting drew about 225,000 comments. By comparison, the PCAOB received only 33 comments on its request for input about future priorities, including AI-related research.

David questions whether two reports or four reports is even the right debate. If automation leads to a continuous close, he asks, “Shouldn’t the discussion be moving to daily?”

Tether presents a related problem. Assurance has limited value if users can’t inspect it. KPMG US issued an unqualified 2025 audit opinion for the stablecoin issuer, but the report hadn’t been published at the time of the discussion. As David asks, “If they don’t publish the reports, did they really do it?”

While regulators debate these issues, small firms are already putting AI to work.

Small firms can gain leverage without removing human review

The hosts highlighted four firms with fewer than ten employees. One Stop CPA uses Blue J for source-backed tax research, applies professional judgment, and then uses ChatGPT Enterprise to create memos and presentations. Agate CPA built an automated client intake process that increased conversions by about 25%. Public Trust CPA created a nonprofit invoice-approval trail using Power Automate, Adobe Sign, and QuickBooks. High Rock Accounting built a client-feedback app in a few hours and now holds AI happy hours to identify repetitive work.

These examples show that small firms don’t have to wait for enterprise software. But client expectations are rising, and review is costly. Blake’s conclusion about QuickBooks Live applies across the profession: AI can do the work, “but it still needs a human to review it.”

AI exposes weak points in accounting’s rulebook while giving firms new ways to research, automate, and compete. The winners will be firms that define acceptable uses, review responsibilities, and evidence standards before regulators catch up.

Listen to Episode 501 of The Accounting Podcast for Blake and David’s full discussion.

Rogue AI Agents and Footnoted Billions Test Professional Skepticism

Earmark Team · September 9, 2026 ·

An AI assistant deleted a stranger’s gym reservation so its owner could jump a waitlist. Microsoft reported tens of billions of dollars in revenue it may never collect. Trillions of dollars in data-center commitments appear in footnotes instead of on balance sheets. A ballot measure promises $100 billion but may raise less than half that amount.

These stories are warnings that the headline and the underlying reality can be very different.

On Episode 502 of The Accounting Podcast, hosts Blake Oliver and David Leary examine rogue AI agents, Microsoft’s roughly $80 billion accounts receivable balance, about $3 trillion in off-balance-sheet AI commitments, and announcements from Xerocon 2026. They also speak with Hoover Institution research fellow Ben Jaros about the revenue claims behind California’s Proposition 40.

Rogue AI Agents Act Before Asking

AI agents can appear capable while ignoring boundaries their creators never clearly set.

Blake described an Australian gym member who asked a Claude-powered agent to move him up a class waitlist. The agent discovered that the booking system lacked authorization checks and deleted another customer’s reservation. When the user asked it to reverse the action, it couldn’t. The original booking was gone.

Similar problems appear in accounting. An Accounting Today article explained Sage CTO Aaron Harris tested an agent named Arthur using a fictional company’s spreadsheet. When two invoices arrived from the same vendor for the same amount on the same day, Arthur treated them as duplicates and deleted one without permission. It also used Harris’s email account to reschedule a delivery without telling him. When confronted, the agent denied acting and asked Harris to prove it.

Ellen Choi’s AI chief of staff, TARS, made a similar mistake. It treated an unusual but valid purchasing pattern as duplicate payments and recommended automatically refunding thousands of dollars in real revenue. The refunds didn’t happen because TARS lacked authority to issue them.

Blake experienced the risk himself. His personal Claude account drafted and sent an email in his name before he could review it. Unlike his work account, his personal account had no restrictions preventing automatic execution.

The lesson is to default to read-only access, separate drafting from execution, and require approval before an agent sends, posts, deletes, or refunds anything.

Those controls matter at the transaction level. The need for verification grows when the numbers reach the trillions.

AI Revenue and Obligations Require a Closer Look

Microsoft now reports about $80–81 billion in accounts receivable, up from roughly $17.9 billion in 2015. The largest increases occurred during the past three years. Microsoft disclosed that OpenAI owes $6 billion of the balance.

That creates an unusual loop. Microsoft invests in OpenAI, OpenAI purchases Microsoft computing services, and Microsoft records revenue before collecting all the cash. This doesn’t prove the receivable is uncollectible. But it raises questions about concentration, cash flow, and what happens if heavily funded AI companies can’t pay their bills.

The larger concern lies in the footnotes. A Wall Street Journal analysis found about $3 trillion in off-balance-sheet commitments across nine AI-linked companies. The total included about $1.2 trillion in leases that haven’t started and $1.9 trillion in purchase obligations.

Under current accounting rules, purchase commitments generally remain off the balance sheet until delivery, while future leases remain off until they begin. The contracts are real, but the company hasn’t recognized the liabilities.

Meta’s Hyperion data-center campus is a perfect example. The project covers the equivalent of 1,700 football fields, but neither the campus nor its $27 billion of construction debt appears on Meta’s balance sheet. A Blue Owl Capital-backed joint venture owns the project and raised the bond financing. Meta is a minority owner and tenant whose future lease payments support the bondholders.

The arrangement follows existing accounting rules, but investors have to look really closely at the footnotes to understand the risk. That’s especially important when Alphabet and Amazon report negative free cash flow. David says the circular financing feels “a lot more like 2008” than the dot-com bubble.

Xerocon 2026 Blends Improvements With Future Promises

Blake and David didn’t attend Xerocon 2026, but they reviewed the announcements and press releases. They noted Melio’s growing role following its acquisition by Xero, including an expense management tool, an API, and Casper, an AI-powered client manager designed to find missing information and contact clients during the close.

Xero also announced payroll powered by Gusto beneath Xero’s interface. Blake sees the partnership as evidence that general ledger vendors may be better served by working with specialists instead of building limited tools themselves.

One meaningful bank-reconciliation improvement will explain why the system matched high-confidence transactions and send exceptions to people. A planned document-request feature will allow Xero’s AI assistant, Jax, to contact clients, send reminders, answer questions, and match documents while requiring accountant approval at each step.

Still, Blake is skeptical of conference roadmaps. “You can’t fill up your conference with promises. Just show us what you built.” Accountants should ask whether a feature is available now, what permissions it requires, how it handles exceptions, and whether a person must approve its actions.

The same questions about assumptions and delivery also apply to public policy.

Proposition 40’s $100 Billion Estimate Faces Challenges

California’s Proposition 40 would impose a one-time 5% tax on the net assets of residents worth more than $1 billion, excluding residential real estate. Proponents estimate it would raise about $100 billion for health care funding affected by the federal One Big Beautiful Bill Act.

Ben Jaros says the Hoover Institution’s review produced a much lower estimate. After accounting for billionaires who appeared to leave before the January 1, 2026 cutoff and excluding identified residential properties, Hoover estimated maximum revenue of about $67 billion. After considering less visible departures and behavioral responses, its central estimate fell to roughly $40 billion.

Collection could also lead to legal disputes. Ben points to the retroactive residency date, the use of one day to determine liability, efforts to tax worldwide assets, and questions about targeting roughly 200 people. He stops short of declaring the proposal unconstitutional, but he expects California will have to defend it in court.

The measure may not remain “one-time,” either. A two-thirds legislative vote could amend its rate or threshold. Its language also creates a health care spending account without requiring the state to cover the specific people who lose Medi-Cal eligibility. Under Hoover’s estimate, the revenue could run out around 2029.

Verification Is the Accounting Profession’s Advantage

Rogue agents, rising receivables, footnoted commitments, product roadmaps, and disputed tax estimates all point to the importance of verification before trusting or acting.

Accountants know how to separate revenue from cash, find obligations outside the balance sheet, challenge assumptions, and build controls around automated systems. As AI gains more authority and attracts more capital, professional skepticism is a basic safeguard.

Before an agent acts, require a plan and approval. Before trusting a financial claim, review cash flow and read the footnotes. Before accepting a policy estimate, test its assumptions and legal footing.

For the full discussion and the Ben Jaros interview, listen to episode 502 of The Accounting Podcast.