
Sign up to save your podcasts
Or


Affairs don’t just ruin relationships; they destroy companies. When the person at the top breaks the rules, the damage doesn’t stay with that person.
Think about what happened at Nvidia this month. After a board investigation into reports that its founder and CEO had been the target of a foreign intelligence effort, he resigned. The stock dropped sharply in a single trading day, putting hundreds of billions of dollars of value in play. And that’s before anyone knew what the investigation would find.
Here’s what C-suite misconduct does. It hits value. It hits culture, because employees watch whether the rules apply to the leader. It hits the business with succession uncertainty, customer questions, regulatory attention, and poor public relations. And it creates legal exposure for the board and officers, who will be asked what they knew, when they knew it, and what they did about it.
The lesson for every company is simple. Your code of conduct, conflicts policy, and security rules must expressly cover the CEO and the board. Your general counsel and compliance officer need a protected line to independent directors, and your board needs a plan for the day the allegation involves the leader.
Compliance programs spend most of their energy on employees. The biggest concentration of risk is often at the top.
The Ethics and Compliance Q and A show is produced by One Stone Creative.
In this episode, Michael Volkov examines the Bureau of Industry and Security's $303,750 penalty against BAE Systems Controls for failing to report boycott-related requests contained in 81 purchase orders from a United Arab Emirates customer between July 2019 and January 2024. The company discovered the language only after BIS added the customer to its Office of Antiboycott Compliance Requestor List, then voluntarily disclosed, admitted the violations, and introduced an automated screening tool. Michael explains why the case turns on the separate duty to report requests even when a company declines to comply, why routine purchase-order boilerplate goes unnoticed, and what compliance teams should do: build antiboycott into export programs, screen incoming documents, check the Requestor List, train commercial staff, create escalation and logging, report on time, conduct lookbacks after triggers, and cover subsidiaries.
For decades, an AML program passed if the boxes were checked. FinCEN now wants to know whether the compliance program actually works.
In April, FinCEN proposed a fundamental rewrite of the AML program rule around a new definition of an effective compliance program with two prongs.
First, establishment.
A risk-based program built on four elements: internal controls, including risk assessment, independent testing, a U.S.-based AML officer, and training, and kept current as your risk changes.
Second, maintenance.
You actually implement it in all material respects.
A beautiful program that lives only on paper fails the second prong.
The proposal also says resources should follow risk.
And it would limit civil enforcement actions against banks that properly establish their programs to significant or systemic failures.
It encourages responsible use of AI, though it leaves criminal exposure for willful violations untouched.
Commentators disagree on how big the change is, and the key phrase “significant or systemic failure” is still undefined.
The comment period has closed, and a final rule is pending, but the direction is clear.
Start measuring outcomes, not activity.
Do your alerts lead to useful reports?
Refresh your risk assessment and govern any AI you rely on.
Effectiveness is a question you should be able to answer today.
The Ethics and Compliance Q and A show is produced by One Stone Creative.
In this episode, Michael Volkov examines the November 12, 2026 resignation of Nvidia CEO Jensen Huang amid an independent board investigation into reports that he was the target of a foreign intelligence effort through a relationship with a Chinese businesswoman, separating what is known from what the investigation has yet to establish. He explains why a CEO's personal conduct becomes a corporate security and governance issue at a company that holds strategic AI technology, the questions the board must answer about who knew what and whether it had executive-level conflict, counterintelligence, and escalation controls, how founder-centered power and culture are tested in a crisis, and what Caremark oversight exposure for directors and officers would actually require a plaintiff to show. He closes with a board checklist: apply codes and conflicts rules expressly to the CEO, give control-function leaders a protected channel to independent directors, consider an independent chair, add insider-threat risk to oversight, rehearse leadership-misconduct protocols, and refresh succession plans.
In this episode, Michael Volkov examines the latest trends in civil antitrust enforcement and litigation and what companies should do now to reduce the risk of government action and private lawsuits. He covers the Antitrust Division's October 2026 message that it prefers early, negotiated remedies but will litigate (illustrated by the abandoned TransDigm–Stellant deal), the focus on household-cost sectors such as healthcare, housing, and food, and the most active civil theories: algorithmic pricing and information exchange after the Third Circuit's Cornish-Adebiyi decision, restrictive contract terms and loyalty rebates, Robinson-Patman revival, and alleged coordination in AI development. He also explains how state attorneys general, new California antitrust laws, local prosecutors, and private class actions are extending enforcement, and closes with a practical checklist: map your exposure, fix high-risk data and contract mechanisms, keep documents consistent with your story, control vendors, plan deals and remedies early, move fast on red flags, and brief the board.
Posting your prices isn’t the same as telling patients the truth about them.
On October 5th of this year, the FTC sent warning letters to 24 of the nation’s largest healthcare services companies, reminding them that patients are entitled to timely, accurate, and complete pricing, especially for scheduled non-emergency care.
Failing to provide it can be an unfair or deceptive practice under the FTC Act.
Here’s the part compliance officers need to hear.
A partial price can be deceptive if an estimate leaves out physician fees or facility fees, or covers only part of the expected care.
The patient may reasonably believe that they’ve been told the total cost.
And the letters say plainly that the CMS price transparency rules are a floor, not a safe harbor.
You can post every file CMS requires and still face FTC liability.
The chairman went further, saying the agency will take noncompliant providers to court and that investigations are already opening.
So review the whole patient journey, not just the posted file: the website estimator, the scheduling call, the financial counselor. Mystery shop a scheduled procedure.
Then compare the quote with the final bill, flag separately billed physician charges, and watch the words “total” and “your cost.”
If it isn’t the total, don’t call it that.
The Ethics and Compliance Q and A show is produced by One Stone Creative.
In this episode, Michael Volkov reviews six trends in criminal antitrust enforcement in 2026 and ten compliance reminders that follow from them. Drawing on the Antitrust Division's statistics, budget submission, and recent speeches by its criminal enforcement chief, he explains that criminal cases filed have rebounded from 9 in fiscal 2023 to 27 in 2025, that prison sentences more than doubled while corporate fines collapsed, and that the docket is centered on government procurement, agriculture, and labor markets, with antitrust charges increasingly paired with fraud. He also covers the Division's shift toward proactive investigations, the "two races" created by corporate leniency and the new whistleblower rewards program, and its record of trial and plea convictions. Michael closes with practical steps: map exposure, train the people who face competitors, use data to find problems first, treat pricing software as a compliance risk, build a fast reporting and leniency-decision path, preserve evidence properly, protect individuals, and keep the board informed.
Most companies have AI in two different rooms. Innovation buys the tools. Compliance writes the policy. Nobody connects it.
Here’s the core idea.
There are two tracks in one program.
Track one: use AI to strengthen compliance in screening, monitoring, third-party diligence, hotline triage, and investigations.
Track two: govern the risk AI creates, from shadow AI and data leakage to unverified output, bias, and deliberate misuse.
And you don’t have to wait for an AI-specific statute.
DOJ’s compliance program guidance already asks how a company governs AI in both its business and its compliance program, and what baseline of human decision-making it uses to judge the tools.
It also asks whether compliance has technology comparable to the business.
OFAC’s 2019 framework covers your screening tools too.
And Amazon and Payoneer paid penalties for weak matching results.
AI isn’t a separate program. It flows through every element of the one you already have.
The Ethics and Compliance Q and A show is produced by One Stone Creative.
In this episode, Michael Volkov examines OFAC's October 27, 2026 settlement with Extreme Networks, which agreed to pay $1,529,685 for nine apparent violations involving exports to MMZA, a Russian defense manufacturer blocked under the 50 percent rule through its ownership by Almaz-Antey. The company had a policy to review every Russia-related dealing, yet OFAC says it ignored red flags that included MMZA email addresses and addresses in purchase orders, a distributor that supplied MMZA's website before switching to a different named end user, and customer portal registrations made under the wrong company name. Michael walks through the penalty calculation, the parallel Justice Department non-prosecution agreement and its "near-miss" characterization of the company's disclosure, and eight practical lessons: treat acquired customer lists as acquired risk, make the 50 percent rule operational, resolve paperwork mismatches with independent evidence, never treat a distributor's statement as a control, follow up on screening hits, connect compliance to what the business knows, watch portals and support channels, and oversee foreign affiliates closely.
You cannot govern AI you cannot see. That’s why the first step is not a policy. It’s an inventory.
Once you’ve found every AI tool in your company, including the shadow ones, sort them into tiers.
High-risk tools that make decisions about people or money, with human review of every output. Moderate risk, customer-facing tools with sampling and disclosure. Low-risk internal productivity tools on approved platforms.
Then build your guardrails: a short, concrete, acceptable use policy that tells employees what they can and cannot paste in.
Vendor diligence that asks who owns the model, where it’s hosted, and whether it trains on your data.
Verification scaled to risk, because in the Avianca case, the court sanctioned the lawyers, not the chatbot.
And an incident plan that covers AI-specific events like a data leak or a deepfake executive call.
Here’s the 90-day plan.
First 30 days: see it. Inventory and brief the board.
Next 30: govern it. Tier and adopt policy.
Final 30: use it. Pilot one compliance AI use case with validation.
Govern first, deploy second.
The Ethics and Compliance Q and A show is produced by One Stone Creative.