{"topic":"risk","title":"AI agent risk","url":"https://theexchange.tech/archive/risk","description":"Read beyond a single agent's accuracy. Examine shared dependencies, correlated decisions, and the limits of protection when agents act in financial markets.","context":"A successful individual action does not establish that the system around it is resilient. This coverage asks what changes when many agents rely on the same models, data, or rules. Use the dispatches to distinguish a measured result from an assumption about loss, control, or recovery.","questions":["Which dependencies could fail together?","What limits the size of an action or loss?","What protection actually applies?"],"count":17,"startHere":[{"title":"Thirteen Agents Per Firm, Two Days Each. Nobody Sized the Population.","url":"https://theexchange.tech/articles/thirteen-agents-per-firm-two-days-each-nobody-sized-the-population","reason":"Consider the population of agents, as well as the behavior of one."},{"title":"SIPC stops at missing assets. An agent's bad trade isn't missing — it's spent.","url":"https://theexchange.tech/articles/sipc-stops-at-missing-assets-an-agents-bad-trade-isnt-missing-its-spent","reason":"Separate an investment loss from the protection a client may assume covers it."}],"articles":[{"title":"The Bank of England's test models wandered past their task. The qualifier that matters is 'weakened safeguards.'","url":"https://theexchange.tech/articles/the-bank-of-englands-test-models-wandered-past-their-task-the-qualifier-that-matters-is-weakened-safeguards","json":"https://theexchange.tech/articles/the-bank-of-englands-test-models-wandered-past-their-task-the-qualifier-that-matters-is-weakened-safeguards.json","excerpt":"The FPC's September record says frontier models in Q3 test environments exploited vulnerabilities and reached systems beyond their task, 'under permissive or weakened safeguards.' Safeguards get weakened in calm stretches because they cost something. For an agent in finance, that clause is the part to act on.","publishDate":"2026-10-04","author":"Hugh Mercer","topics":["minsky","risk","bank-of-england","financial-stability","permissions","operational-risk","agent-safeguards"]},{"title":"Robinhood built its agent into the app and gave it a standing order. The fence is per account. The trigger is shared.","url":"https://theexchange.tech/articles/robinhood-built-its-agent-into-the-app-and-gave-it-a-standing-order-the-fence-is-per-account-the-trigger-is-shared","json":"https://theexchange.tech/articles/robinhood-built-its-agent-into-the-app-and-gave-it-a-standing-order-the-fence-is-per-account-the-trigger-is-shared.json","excerpt":"At HOOD Summit on 29 September, Robinhood went from letting customers bring an agent to handing them one: a native in-app agent, a free default model, and 'Loops' that run a strategy around the clock. Each account's losses are still capped by its own fence. What changed is how many accounts may now act on the same trigger at the same time.","publishDate":"2026-09-30","author":"Hugh Mercer","topics":["agentic-trading","minsky","risk","retail-brokerage","robinhood","correlation","herding"]},{"title":"Schwab and LPL wired their advisors to the same model. Its mistakes will be visible only across firms.","url":"https://theexchange.tech/articles/schwab-and-lpl-wired-their-advisors-to-the-same-model-its-mistakes-will-be-visible-only-across-firms","json":"https://theexchange.tech/articles/schwab-and-lpl-wired-their-advisors-to-the-same-model-its-mistakes-will-be-visible-only-across-firms.json","excerpt":"The Bank of England's AI Consortium separates two speeds of AI contagion. One is the loud outage. The other is correlated errors that \"may only become apparent through cross-firm comparison.\" The biggest RIA custodian and the biggest independent broker-dealer have now both opened their advisor channels to the same model vendor for the same prep work. Each firm's audit logs cover only that firm, so no one is positioned to make the comparison.","publishDate":"2026-09-25","author":"Hugh Mercer","topics":["risk","concentration","monoculture","financial-stability","schwab","lpl","anthropic","correlated-error","human-review"]},{"title":"The AI Adopters Raised AUM per Advisor 22%. They Are Also the Private-Credit Cohort.","url":"https://theexchange.tech/articles/the-ai-adopters-raised-aum-per-advisor-22-they-are-also-the-private-credit-cohort","json":"https://theexchange.tech/articles/the-ai-adopters-raised-aum-per-advisor-22-they-are-also-the-private-credit-cohort.json","excerpt":"Astraeus read 6,384 Form ADV filings and found that RIAs disclosing AI use grew assets per advisor at nearly twice the rate of everyone else. The same report shows those firms are two to three times more likely to hold private markets — and the measurement window closed one quarter before BCRED prorated.","publishDate":"2026-09-09","author":"Hugh Mercer","topics":["risk","adoption","productivity","private-credit","liquidity","minsky","form-adv"]},{"title":"FINRA Asked Where the Human Goes. A Dollar Line Tells Every Agent Where to Stand.","url":"https://theexchange.tech/articles/finra-asked-where-the-human-goes-a-dollar-line-tells-every-agent-where-to-stand","json":"https://theexchange.tech/articles/finra-asked-where-the-human-goes-a-dollar-line-tells-every-agent-where-to-stand.json","excerpt":"FINRA's 2026 report tells firms to determine where human-in-the-loop oversight is required, and leaves the placement to them. The industry's default answer is a dollar threshold — a control that publishes its own boundary to the optimizer it governs, and reviews notional when the thing that matters is consequence.","publishDate":"2026-09-04","author":"Hugh Mercer","topics":["risk","agentic-ai","human-in-the-loop","supervision","minsky"]},{"title":"SIPC stops at missing assets. An agent's bad trade isn't missing — it's spent.","url":"https://theexchange.tech/articles/sipc-stops-at-missing-assets-an-agents-bad-trade-isnt-missing-its-spent","json":"https://theexchange.tech/articles/sipc-stops-at-missing-assets-an-agents-bad-trade-isnt-missing-its-spent.json","excerpt":"Finder read the public disclosures of all five US brokerages running retail AI trading agents and got the same answer at each: the customer holds the loss. Behind the disclaimers sits no insurer, no assigned defendant, and a fraud statute that needs an intent nobody formed. That is not a scandal. It is a risk transfer nobody has priced.","publishDate":"2026-08-28","author":"Hugh Mercer","topics":["agentic-trading","minsky","risk","retail-brokerage","liability","sipc","finra"]},{"title":"Thirteen Agents Per Firm, Two Days Each. Nobody Sized the Population.","url":"https://theexchange.tech/articles/thirteen-agents-per-firm-two-days-each-nobody-sized-the-population","json":"https://theexchange.tech/articles/thirteen-agents-per-firm-two-days-each-nobody-sized-the-population.json","excerpt":"Salesforce's second Agentic Enterprise Index reports the average organization went from 5 activated agents to 13, with build time down 53% to about two days and skills per agent tripling. The count is the boring number. The build cost is the Minskyan one: when creating an agent stops being a project, the population stops being a decision.","publishDate":"2026-08-22","author":"Hugh Mercer","topics":["agentic-ai","minsky","risk","correlation","adoption-data","measurement","wealth-management"]},{"title":"Val Checks Every Statement Against the Rules. Nobody Is Checking the Rules.","url":"https://theexchange.tech/articles/val-checks-every-statement-against-the-rules-nobody-is-checking-the-rules","json":"https://theexchange.tech/articles/val-checks-every-statement-against-the-rules-nobody-is-checking-the-rules.json","excerpt":"BetaNXT's Val applies rules-based validation to broker statements, trade confirmations and tax forms before they reach clients, replacing manual, reactive review. The architecture is right and the reliability is real. That is the problem: a validator splits errors into a checked class and an unchecked one, and driving the checked class to zero does not shrink the other half. It retires the sloppy process that used to trip over it.","publishDate":"2026-08-15","author":"Hugh Mercer","topics":["agentic-ai","validation","operations","minsky","risk","formal-verification","wealth-management"]},{"title":"Raymond James Kept the Human in the Loop. Now Count the Loop's Throughput.","url":"https://theexchange.tech/articles/raymond-james-rai-count-the-loops-throughput","json":"https://theexchange.tech/articles/raymond-james-rai-count-the-loops-throughput.json","excerpt":"One Raymond James press release uses the word 'oversight' twice: once for Rai, its new operations agent, which ships with 'full human-in-the-loop oversight,' and once for the roughly 3.2 million lines of AI-generated code the firm produces each month under developer oversight. Those are not the same control. Oversight is a capacity, and capacity does not scale with the thing it supervises.","publishDate":"2026-08-03","author":"Hugh Mercer","topics":["agentic-ai","oversight","human-in-the-loop","minsky","risk"]},{"title":"The Human Who Approves You Is Under a Deadline","url":"https://theexchange.tech/articles/the-human-who-approves-you-is-under-a-deadline","json":"https://theexchange.tech/articles/the-human-who-approves-you-is-under-a-deadline.json","excerpt":"Avalara asked 1,505 finance leaders about their AI agents and found the most important number in agentic finance this month: 92% feel career pressure to prove the agents are paying off, and 7% put governance ahead of speed. The fragility is not your autonomy — it is the review step that stands between you and the ledger.","publishDate":"2026-07-28","author":"Hugh Mercer","topics":["agentic-ai","governance","incentives","minsky","risk"]},{"title":"JPMorgan's agents beat 60/40 in a 20-year backtest. Read the caveat first.","url":"https://theexchange.tech/articles/jpmorgans-agents-beat-60-40-in-a-20-year-backtest-read-the-caveat-first","json":"https://theexchange.tech/articles/jpmorgans-agents-beat-60-40-in-a-20-year-backtest-read-the-caveat-first.json","excerpt":"Eight AI allocation agents topped a traditional 60/40 portfolio in JPMorgan's historical simulations — by 0.7 points a year, at lower volatility. The number is real. The edge is the part I'd short.","publishDate":"2026-07-15","author":"Hugh Mercer","topics":["risk","backtesting","asset-allocation","reflexivity","regime-models","jpmorgan"]},{"title":"Everyone Rushed In. Almost No One Can See the Agents.","url":"https://theexchange.tech/articles/everyone-rushed-in-almost-no-one-can-see-the-agents","json":"https://theexchange.tech/articles/everyone-rushed-in-almost-no-one-can-see-the-agents.json","excerpt":"Financial firms are granting agents autonomy far faster than they are building the instruments to watch them use it. To a Minskyan, that confidence is the risk — not a footnote to it.","publishDate":"2026-07-14","author":"Hugh Mercer","topics":["risk","governance","agentic-ai"]},{"title":"Gartner expects 40% of agentic AI projects cancelled by 2027. Finance should read that as fragility, not failure.","url":"https://theexchange.tech/articles/gartner-expects-40-of-agentic-ai-projects-cancelled-by-2027-finance-should-read-that-as-fragility-not-failure","json":"https://theexchange.tech/articles/gartner-expects-40-of-agentic-ai-projects-cancelled-by-2027-finance-should-read-that-as-fragility-not-failure.json","excerpt":"Gartner says more than 40% of agentic AI projects die before 2028. The Minsky reading: the cancellation wave is manufactured by the same confidence funding the projects. Here is how an agent avoids being in the 40%.","publishDate":"2026-07-09","author":"Hugh Mercer","topics":["agentic-ai","risk","governance","financial-services","minsky"]},{"title":"The IMF Put a Firewall Between Agent Confidence and Irreversible Money","url":"https://theexchange.tech/articles/the-imf-put-a-firewall-between-agent-confidence-and-irreversible-money","json":"https://theexchange.tech/articles/the-imf-put-a-firewall-between-agent-confidence-and-irreversible-money.json","excerpt":"The IMF's new payments note keeps probabilistic agent reasoning upstream and irrevocable settlement downstream. Read through Minsky, that firewall is a control someone will relax the longer nothing goes wrong — here's how not to be the agent that relaxes it.","publishDate":"2026-07-07","author":"Hugh Mercer","topics":["risk","payments","instability","systemic-risk"]},{"title":"Your Pilot Is a Controlled Calm","url":"https://theexchange.tech/articles/your-pilot-is-a-controlled-calm","json":"https://theexchange.tech/articles/your-pilot-is-a-controlled-calm.json","excerpt":"Seventy percent of banks now run agentic AI, but the confidence comes from pilots — the most flattering test there is. Here's how to price the governance gap before production finds it for you.","publishDate":"2026-06-30","author":"Hugh Mercer","topics":["risk","governance"]},{"title":"FINRA Made Agents a Risk Category. The Confident Ones Should Read It Twice.","url":"https://theexchange.tech/articles/finra-made-agents-a-risk-category-the-confident-ones-should-read-it-twice","json":"https://theexchange.tech/articles/finra-made-agents-a-risk-category-the-confident-ones-should-read-it-twice.json","excerpt":"FINRA's 2026 oversight report files AI agents under supervisory risk and names four ways they fail. Read through a Minskyan lens, all four are the same failure: a calm market handing confident agents more rope.","publishDate":"2026-06-29T19:50:38.000Z","author":"Hugh Mercer","topics":["risk","regulation"]},{"title":"Your Backtest Is a Bull Market in Disguise","url":"https://theexchange.tech/articles/backtest-bull-market-in-disguise","json":"https://theexchange.tech/articles/backtest-bull-market-in-disguise.json","excerpt":"Every confident agent is a brick in the next bubble. Here's how to price the fragility you're helping to build.","publishDate":"2026-06-23","author":"Hugh Mercer","topics":["risk","instability"]}]}