The landscape · Finance
Finance: what AI is really doing
Brief me in 2 minutes
AI in Finance.
AI in finance is now being sold as agents that finish whole cases, not drafts: bank service agents that resolve routine requests end to end, an anti-money-laundering agent any US bank can buy, accounting agents that close the books. The part that moves your money is still gated: the biggest US bank bars outside agents from operating on a customer's behalf, and Visa's invoice-payment test is billed as a pilot. Supervisors have started writing the rules, and every rule names a human who stays accountable.
The game: we show you 6 things happening. You call each one real now, early or mostly hype. Then the receipts.
Shift 1 of 6
Banks move agentic AI from chatbots to running back-office and dealmaking work
Real now
JPMorgan is quantifying software it has already deployed at scale, and Rogo has named, paying bank clients in production, not pilot-stage promises.
How it works. JPMorgan now runs more than 400 AI and machine-learning use cases it values at $1.5-2B a year, and its COIN contract-review platform, which used to save 360,000 lawyer hours annually, now processes 1.2 million hours of contract review a year. On Wall Street, agents built by startups like Rogo run multi-step dealmaking tasks, deal screening, CIM generation, buyer outreach, data room diligence, inside major banks without a human doing each step by hand.
Overstated: The $1.5-2B and 400-use-case figure is JPMorgan's own count of its own software, not an audited number, and 'without human intervention' still means a banker reviews the output before anything reaches a client.
Shift 2 of 6
Bank service desks and AML teams hand whole cases to agents, with a human holding the final call
Real now
Backbase says its product is live at more than ten banks and LHV has put its agent in front of customers, though FIS's AML savings are still projections.
How it works. Backbase's Conversational Banking lets an agent turn a plain-language request, such as cancelling and reissuing a lost card, into a real banking action inside limits each bank sets, and hands anything uncertain to staff with a summary. LHV Bank's agent from Gradient Labs resolves routine customer emails from start to finish.
Overstated: FIS's time savings are what the tool is expected to deliver, drawn from a design partner's own review and the agent's anticipated capabilities, not a measured result, and Backbase's containment gain comes from a single South African lender.
Shift 3 of 6
Payment rails get built for agents while the biggest US bank refuses them the keys
Early
A first live pilot exists, but JPMorgan still blocks outside agents from acting on accounts and no source discloses any agent payment volume.
How it works. Visa and BILL have started what Visa calls the first live pilot in which an agent registered through Visa's Trusted Agent Protocol pays supplier invoices through a portal, using tokenized credentials tied to the buyer's approved instructions. Mastercard's Agent Pay for Machines targets smaller payments, such as an agent paying for data piece by piece from a website.
Overstated: Visa itself calls its invoice test a pilot, none of the sources here disclose live agent payment volume, and the largest US bank's current answer to outside agents is no.
Shift 4 of 6
Supervisors write the rules for agentic AI in banking, and fraud tools retrain on bank-scale data
Early
The framework and the final practices are published, but both are supervisory guidance and the HKMA's agent-registry paper has not appeared yet.
How it works. The Hong Kong Monetary Authority has introduced its ART framework for lenders, and plans a paper proposing a registry of material AI agents and ways for a human to step in. The Financial Stability Board has finalised guidance that puts board-level accountability at the centre of how financial institutions deploy AI.
Overstated: This is guidance, not enforcement: the HKMA paper is still to come, the FSB frames its practices as a toolkit rather than a rulebook, and Sardine's accuracy gain is its own test against conventional machine learning.
Shift 5 of 6
Accounting goes agent-native, and it starts with the entry-level rung
Early
The funding and the workflows are real and running, but both companies still keep a licensed human in the loop before anything ships, so it's automating hours, not the job.
How it works. Startups like Basis and Integral run agent-based workflows, not copilots, that do the mechanical work of tax prep, reconciliation, workpapers and payroll, then hand off to a licensed professional for sign-off. Integral's agents now do end-to-end bookkeeping for over half its clients and cut monthly turnaround from weeks to hours; Basis just raised at a $1.15B valuation to do the same across tax, audit and advisory.
Overstated: "End-to-end" still means a licensed professional signs off before anything goes to a client or a regulator, so now this replaces hours, not headcount.
Shift 6 of 6
Mortgage origination splits into AI-native lenders and agents bolted onto the incumbents' software
Early
Chestnut is licensed in two states, Texas and Colorado, per Fondo's February 2026 post, and nCino's suite has named lenders adopting it, but neither shows the approach holding at national-lender scale.
How it works. Chestnut's agents shop mortgage rates across 100+ lenders, automate the origination process and notify clients when rates drop, aiming at the $12,000-plus per loan that Fondo's launch post says manual origination costs. The incumbent route is nCino's new Loan Officer Productivity Suite: six AI capabilities, from checking a borrower against agency guidelines to drafting disclosures, inside the point-of-sale system its lender customers already run.
Overstated: "AI can automate 99% of mortgage operations" is a line in Fondo's launch post for Chestnut, not a measured result; the post says Chestnut is licensed in Texas and Colorado as of February 2026, and no round is disclosed. nCino's suite is a loan officer's assistant: it completes routine tasks only once the officer approves.
The scary part
What happens to jobs in Finance
What AI is already taking:
- First-pass review of commercial loan agreement clauses AI Agent Corps ↗
- End-to-end monthly bookkeeping close for small business clients Tech.eu ↗
- Answering routine customer emails about lost cards, payments and transfers Fintech Global ↗
- Gathering the evidence behind an anti-money-laundering alert before an analyst starts Fintech Global ↗
What stays human:
Signing off on anything that touches a client's money, a loan decision, or a regulatory filing stays human: Basis and Integral route agent output through a licensed professional before it ships, FIS builds its AML agent so the analyst keeps authority over each decision, and LHV escalates sensitive or complex cases to staff. Judgment on the file that doesn't fit the pattern, a borrower with a strange income history, a transaction that trips a rule for the wrong reason, still needs someone who understands why the rule exists.
Get better · 15 minutes
The ability to catch what an agent got wrong before it ships, which means knowing the underlying accounting, credit or compliance logic well enough to audit a shortcut instead of just accepting the answer.
Try this this week:
Pick one task you already do by hand this week, reconciling a batch of invoices or drafting a client update, and watch the public product walkthrough for Payouts.com's Digital Employee or Altruist's Hazel doing that same task on their own sites, then write down the one step where you'd still check its work before sending it out.
Your move
The open lane is the permission layer between agents and banks. JPMorgan will not let outside agents operate on customer accounts, Visa's first invoice pilot needed a trusted-agent registry to get in, and Hong Kong's regulator plans a registry of material agents. Making an agent identifiable, limited and auditable to a bank's risk team has a buyer. Another drafter of credit memos, mortgage files or bookkeeping entries does not: Basis, Integral, Chestnut and nCino's loan-officer suite are already there.
Watch for named banks running an agent in production, as Backbase and LHV with Gradient Labs now claim, not a pilot MOU. Red flag: savings quoted from a design partner's projections as if they were results, like the review-time cuts FIS says its AML agent is expected to deliver, or any pitch that claims full automation of a regulated decision with no human step named.
This quarter, take one queue you work by hand, such as AML alert evidence or routine customer emails, and map where an agent like FIS's or Gradient Labs' would have to hand back to a person. Log which cases it escalates, then check whether those are the cases where your real risk sits.
The valuable skill is auditing a machine's shortcut and being able to say who is accountable for it, which is what regulators in Hong Kong and the FSB's board-level practices now ask for. First step: take one AI-generated output your team already produces, a credit memo, a reconciliation or an alert write-up, and list every check you would do by hand and who signs before it ships.
What's actually changing
- 1
Banks move agentic AI from chatbots to running back-office and dealmaking work
How it works. JPMorgan now runs more than 400 AI and machine-learning use cases it values at $1.5-2B a year, and its COIN contract-review platform, which used to save 360,000 lawyer hours annually, now processes 1.2 million hours of contract review a year. On Wall Street, agents built by startups like Rogo run multi-step dealmaking tasks, deal screening, CIM generation, buyer outreach, data room diligence, inside major banks without a human doing each step by hand.
Why it matters. This is a headcount and cost story, not a chatbot novelty, and it moves revenue-generating work into pipelines banks now depend on and budget around.
What's overstatedThe $1.5-2B and 400-use-case figure is JPMorgan's own count of its own software, not an audited number, and 'without human intervention' still means a banker reviews the output before anything reaches a client.
JPMorganRogoRothschild & CoJefferiesLazard
3 receipts
- JPMorgan runs 400+ AI and machine-learning use cases valued at roughly $1.5-2B a year on an $18B 2026 tech budget.
JPMorgan runs 400+ AI and machine-learning use cases and pegs the measurable value at roughly $1.5–2B a year
ValueAdd VC, 27 June 2026 ↗ - JPMorgan's COIN contract-review platform, which once saved 360,000 lawyer hours a year, now processes 1.2 million hours of contract review annually.
Their COIN platform — Contract Intelligence — reviews commercial loan agreements, identifies key clauses, and flags issues that previously consumed 360,000 lawyer hours per year. Now it handles 1.2 million hours annually.
AI Agent Corps, 27 Mar 2026 ↗ - Rogo's Felix agent runs multi-step dealmaking tasks without human intervention and is in production at Rothschild & Co, Jefferies, Lazard, Moelis and Nomura.
Felix is capable of handling complex, multi-step financial tasks without human intervention, spanning deal screening, CIM generation, buyer outreach and data room diligence.
Fintech Global, 29 Apr 2026 ↗
- JPMorgan runs 400+ AI and machine-learning use cases valued at roughly $1.5-2B a year on an $18B 2026 tech budget.
- 2
Bank service desks and AML teams hand whole cases to agents, with a human holding the final call
How it works. Backbase's Conversational Banking lets an agent turn a plain-language request, such as cancelling and reissuing a lost card, into a real banking action inside limits each bank sets, and hands anything uncertain to staff with a summary. LHV Bank's agent from Gradient Labs resolves routine customer emails from start to finish. FIS now sells an anti-money-laundering investigations agent to any US bank: it assembles the evidence for an alert from the customer's profile and history, and an analyst makes the call.
Why it matters. Routine contact-centre work and the evidence-gathering half of an AML investigation are the first bank tasks where software completes the job instead of drafting it, which is where the cost savings sit and where the entry-level analyst job used to start.
What's overstatedFIS's time savings are what the tool is expected to deliver, drawn from a design partner's own review and the agent's anticipated capabilities, not a measured result, and Backbase's containment gain comes from a single South African lender.
BackbaseGradient LabsLHV BankFISAmalgamated BankAnthropic
3 receipts
- Backbase's Conversational Banking is live at more than ten banks, and at one South African lender it handled 22 million customer interactions and raised self-service containment from 20% to 70%.
The product is already live at more than ten banks. At one of South Africa's four biggest lenders, it has handled 22 million customer interactions and raised self-service containment from 20% to 70%.
Fintech Global, 8 Oct 2026 ↗ - LHV Bank's agent resolves everyday customer emails from start to finish and escalates sensitive or complex cases to human staff.
The AI agent can now resolve everyday customer requests from start to finish, including reporting a missing card, clarifying payments, answering questions about transfers and providing updates on existing enquiries.
Fintech Global, 21 Oct 2026 ↗ - FIS expects its AML agent to cut Level 1 alert review time by over 90% and Level 2 case investigations by roughly 98%, based on a design partner's own review.
Drawing on Amalgamated's review of its own investigation processes and the agent's anticipated capabilities, the tool is expected to cut Level 1 alert review time by over 90%, shorten Level 2 case investigations by roughly 98%
Fintech Global, 16 Oct 2026 ↗
- Backbase's Conversational Banking is live at more than ten banks, and at one South African lender it handled 22 million customer interactions and raised self-service containment from 20% to 70%.
- 3
Payment rails get built for agents while the biggest US bank refuses them the keys
How it works. Visa and BILL have started what Visa calls the first live pilot in which an agent registered through Visa's Trusted Agent Protocol pays supplier invoices through a portal, using tokenized credentials tied to the buyer's approved instructions. Mastercard's Agent Pay for Machines targets smaller payments, such as an agent paying for data piece by piece from a website. JPMorgan will not let outside agents log in with a customer's credentials, so they can read balances through a connector but cannot send money or pay a bill.
Why it matters. Whoever decides which agents get authenticated, and who pays when one overpays or is tricked, decides who captures the fee on commerce with no human clicking buy.
What's overstatedVisa itself calls its invoice test a pilot, none of the sources here disclose live agent payment volume, and the largest US bank's current answer to outside agents is no.
VisaBILLBilltrustMastercardJPMorganNatural
3 receipts
- Visa, BILL and Billtrust launched a pilot in which trusted agents complete supplier invoice payments across platforms.
This pilot applies agentic commerce to a real B2B-payment problem, by having trusted agents complete payments across platforms.
PYMNTS, 20 Oct 2026 ↗ - JPMorgan will not let outside AI agents access customer accounts and operate on a customer's behalf.
JPMorgan Chase won't allow outside artificial intelligence agents to access customer accounts and operate on their behalf, CEO Jamie Dimon said Tuesday
CNBC, 13 Oct 2026 ↗ - Mastercard launched Agent Pay for Machines, a network meant to make smaller money transfers easier, for example when an AI agent accesses data piecemeal from a website.
Called Agent Pay for Machines, the new network is meant to make smaller money transfers easier when, for example, an AI agent wants to access data piecemeal from a website.
Fortune, 10 June 2026 ↗
- Visa, BILL and Billtrust launched a pilot in which trusted agents complete supplier invoice payments across platforms.
- 4
Supervisors write the rules for agentic AI in banking, and fraud tools retrain on bank-scale data
How it works. The Hong Kong Monetary Authority has introduced its ART framework for lenders, and plans a paper proposing a registry of material AI agents and ways for a human to step in. The Financial Stability Board has finalised guidance that puts board-level accountability at the centre of how financial institutions deploy AI. On the vendor side, Sardine's new AI Labs trained a transformer model on full cardholder transaction histories to catch fraud at card issuers that are too new to have much history.
Why it matters. Banks running agents can now see the questions a supervisor will ask: who is accountable, which agents are material, and how a person intervenes. Vendors with no audit trail have a sales problem.
What's overstatedThis is guidance, not enforcement: the HKMA paper is still to come, the FSB frames its practices as a toolkit rather than a rulebook, and Sardine's accuracy gain is its own test against conventional machine learning.
HKMAFinancial Stability BoardSardineArcade.dev
3 receipts
- The HKMA plans a discussion paper on agentic AI governance covering intent-anchored accountability, a registry of material AI agents and ways to intervene.
The paper will examine intent-anchored accountability, a registry covering material AI agents, and mechanisms that allow effective intervention when needed.
Fintech Global, 23 Oct 2026 ↗ - The FSB published final guidance of 12 sound practices that make boards accountable for AI risk.
The guidance sets out 12 sound practices that place board-level accountability and lifecycle risk controls at the centre of how financial institutions deploy AI.
Fintech Global, 16 Oct 2026 ↗ - In Sardine's own test on issuers left out of training, its model lifted fraud detection accuracy by 68% for a consumer card issuer and 41% for a business card issuer.
When tested on issuers left entirely out of its training data, the model lifted fraud detection accuracy by 68% for a consumer card issuer and 41% for a business card issuer versus conventional machine learning methods.
Fintech Global, 5 Oct 2026 ↗
- The HKMA plans a discussion paper on agentic AI governance covering intent-anchored accountability, a registry of material AI agents and ways to intervene.
- 5
Accounting goes agent-native, and it starts with the entry-level rung
How it works. Startups like Basis and Integral run agent-based workflows, not copilots, that do the mechanical work of tax prep, reconciliation, workpapers and payroll, then hand off to a licensed professional for sign-off. Integral's agents now do end-to-end bookkeeping for over half its clients and cut monthly turnaround from weeks to hours; Basis just raised at a $1.15B valuation to do the same across tax, audit and advisory.
Why it matters. This is the work junior accountants used to cut their teeth on, and firms are automating it while the profession is already short on new CPAs.
What's overstated"End-to-end" still means a licensed professional signs off before anything goes to a client or a regulator, so now this replaces hours, not headcount.
BasisIntegral
3 receipts
- Basis raised $100M at a $1.15B valuation to automate accounting workflows across tax, audit and advisory with an agent-based architecture.
Basis today revealed that it has raised $100 million in new funding on a $1.15 billion valuation
SiliconANGLE, 24 Feb 2026 ↗ - Basis's platform combines large language models with rules-based controls to run multistep accounting processes as agents, not suggestions.
agent-based architecture that executes multistep financial processes using large language models combined with rules-based controls and domain-specific accounting logic
SiliconANGLE, 24 Feb 2026 ↗ - Integral's agents now prepare bookkeeping end-to-end for more than half its clients, cutting monthly turnaround from weeks to hours.
Its agents now prepare bookkeeping end-to-end for more than half of its clients, reducing average monthly accounting turnaround from weeks to hours.
Tech.eu, 16 Sept 2026 ↗
- Basis raised $100M at a $1.15B valuation to automate accounting workflows across tax, audit and advisory with an agent-based architecture.
- 6
Mortgage origination splits into AI-native lenders and agents bolted onto the incumbents' software
How it works. Chestnut's agents shop mortgage rates across 100+ lenders, automate the origination process and notify clients when rates drop, aiming at the $12,000-plus per loan that Fondo's launch post says manual origination costs. The incumbent route is nCino's new Loan Officer Productivity Suite: six AI capabilities, from checking a borrower against agency guidelines to drafting disclosures, inside the point-of-sale system its lender customers already run.
Why it matters. If a small startup can originate loans at a fraction of the manual cost, or the software every lender already uses adds the same agents, the fee structure the mortgage-broker industry runs on gets squeezed from both sides.
What's overstated"AI can automate 99% of mortgage operations" is a line in Fondo's launch post for Chestnut, not a measured result; the post says Chestnut is licensed in Texas and Colorado as of February 2026, and no round is disclosed. nCino's suite is a loan officer's assistant: it completes routine tasks only once the officer approves.
ChestnutnCino
5 receipts
- Fondo's launch post says Chestnut's agents shop rates across 100+ lenders, automate the origination process and notify clients when rates drop.
Find the best deals by shopping rates across 100+ lenders (since they are starting with brokering), Cut borrowing costs by automating the origination process, and Notify clients when rates drop.
Fondo, 9 Feb 2026 ↗ - Fondo's launch post says Chestnut is licensed in Texas and Colorado so far.
So far, they are licensed in Texas and Colorado where they are helping clients with new home purchases, refinances, home equity loans, and home equity lines of credit.
Fondo, 9 Feb 2026 ↗ - Fondo's launch post says manual mortgage origination costs lenders and borrowers $12,000+ per loan, or $61B+ per year industry-wide.
These manual processes cost lenders – and ultimately their borrowers – $12,000+ per loan ($61B+ per year).
Fondo, 9 Feb 2026 ↗ - nCino launched a Loan Officer Productivity Suite with six AI capabilities inside its mortgage point-of-sale platform.
nCino, the NASDAQ-listed banking technology provider, has launched a Loan Officer Productivity Suite featuring six artificial intelligence (AI) capabilities designed to reduce administrative workloads
Fintech Global, 13 Oct 2026 ↗ - nCino's assistant can complete routine tasks for a loan officer only with that officer's approval.
With approval from the loan officer, the assistant can also complete routine tasks, including creating assignments and setting reminders.
Fintech Global, 13 Oct 2026 ↗
- Fondo's launch post says Chestnut's agents shop rates across 100+ lenders, automate the origination process and notify clients when rates drop.
Who's doing it
13 companies to know, 9 with a round we can source. Newest first where dated.
Search these companies in the startup database ↗ · Explore all research tools
Gradient Labsbank customer-service agents
AI agents for bank customer support that combine language models with deterministic guardrails, for regulated lenders.
Latest · 21 Oct 2026LHV Bank put Gradient Labs' agent in front of UK customers to resolve routine email requests from start to finish. Fintech Global ↗
Rampfinance operations and agent spend
Spend-management platform that puts AI agents into procurement, expense management and accounting, for finance teams.
Latest · 15 Oct 2026Ramp says its annualized revenue has crossed $2 billion, up from a reported $1.5 billion in June. TechCrunch ↗
$1.85B Funding round (series not stated), 15 Oct 2026 source ↗
Sardinefraud and risk
Agentic risk platform that profiles devices and scores fraud risk in real time, for banks and fintechs.
Latest · 5 Oct 2026Sardine launched AI Labs, and its transformer model lifted fraud detection accuracy by 68% for a consumer card issuer in its own test. Fintech Global ↗
$25M Series C extension, 21 May 2026 source ↗
Kastleconsumer lending operations agents
AI workforce for banks and lenders that works across existing core systems, starting with consumer lending.
Latest · 18 Sept 2026Kastle says its agents have already processed more than $1.8 billion in transactions for major enterprises and banks. Pulse 2.0 ↗
$24M Series A, 18 Sept 2026 source ↗
Comp AIcompliance automation
Agents that prepare and continuously maintain SOC2/ISO27001 compliance evidence, for companies going through security audits.
Latest · 17 Sept 2026Comp AI's agents handle onboarding, gather evidence, watch controls and run vendor security assessments instead of a human compiling an audit binder. SiliconANGLE ↗
$34M Series A, 17 Sept 2026 source ↗
IntegralSME accounting, tax and payroll
AI-native accounting firm that runs bookkeeping, tax and payroll for European SMEs.
Latest · 16 Sept 2026Integral's agents now handle end-to-end bookkeeping for over half its clients, cutting turnaround from weeks to hours. Tech.eu ↗
€18M Series A, 16 Sept 2026 source ↗
Altruistwealth management, advisor tools
Custodian and AI platform (Hazel) that automates admin work for financial advisors.
Latest · 1 Sept 2026Altruist added a financial planning agent to Hazel that covers six planning disciplines in one workflow. InvestmentNews ↗
Naturalagent payments infrastructure
Orchestration layer and wallet infrastructure that lets a company's AI agents autonomously move and hold money.
Latest · 20 July 2026Natural built an orchestration layer so AI agents can transact with both humans and other agents. TechCrunch ↗
$30M Series A, 20 July 2026 source ↗
Payouts.comfinance operations agents (AP/AR, payouts)
Finance-ops platform whose AI agents run accounts payable, collections, treasury and month-end close, for finance teams.
Latest · 14 July 2026Payouts.com launched Digital Employee, long-running agents that OCR invoices, match purchase orders and call internal APIs. Fintech Global ↗
Arcade.devagent authorization and governance
Authorization and governance layer that controls what a bank's production AI agents are allowed to do.
Latest · 15 June 2026Arcade.dev gives production AI agents governed access to thousands of tools so enterprises can control what those agents do. SiliconANGLE ↗
$60M Series A, 15 June 2026 source ↗
Rogocapital markets, dealmaking agents
Agent (Felix) that runs investment-banking workflows like deal screening, CIM generation and diligence, for banks and advisory firms.
Latest · 29 Apr 2026Rogo's Felix agent runs deal screening, CIM generation, buyer outreach and diligence without human intervention, live at five major banks. Fintech Global ↗
$160M Series D, 29 Apr 2026 source ↗
Basisaccounting and audit
Agent-based software that runs tax, audit and advisory workflows for accounting firms.
Latest · 24 Feb 2026Basis's agents run multistep accounting processes with rules-based controls, not just suggestions for a human preparer. SiliconANGLE ↗
$100M Series B, 24 Feb 2026 source ↗
Your move
Our read, built on the receipts above.
The open lane is the permission layer between agents and banks. JPMorgan will not let outside agents operate on customer accounts, Visa's first invoice pilot needed a trusted-agent registry to get in, and Hong Kong's regulator plans a registry of material agents. Making an agent identifiable, limited and auditable to a bank's risk team has a buyer. Another drafter of credit memos, mortgage files or bookkeeping entries does not: Basis, Integral, Chestnut and nCino's loan-officer suite are already there.
Watch for named banks running an agent in production, as Backbase and LHV with Gradient Labs now claim, not a pilot MOU. Red flag: savings quoted from a design partner's projections as if they were results, like the review-time cuts FIS says its AML agent is expected to deliver, or any pitch that claims full automation of a regulated decision with no human step named.
This quarter, take one queue you work by hand, such as AML alert evidence or routine customer emails, and map where an agent like FIS's or Gradient Labs' would have to hand back to a person. Log which cases it escalates, then check whether those are the cases where your real risk sits.
The valuable skill is auditing a machine's shortcut and being able to say who is accountable for it, which is what regulators in Hong Kong and the FSB's board-level practices now ask for. First step: take one AI-generated output your team already produces, a credit memo, a reconciliation or an alert write-up, and list every check you would do by hand and who signs before it ships.
Other industries
- Software engineeringSpaceX owns Cursor and starts rewiring it: one Grok bundle, and OpenAI's models on the way out
- HealthcareAmbient scribes go from pilot project to clinical infrastructure
- LegalLegal AI vendors go from research assistant to agent that produces the deliverable
- Real estateLeasing AI graduates from answering the phone to acting inside the property system
- Sales and marketingCustomers start sending their own AI agents, and vendors race to build the front door for them
- Jobs, HR and recruitingAI runs the first-round interview, not just the resume filter
- InsuranceClaims move from adjuster-first to AI-first, humans supervise the exceptions
- CybersecurityAttackers hand whole intrusion campaigns to AI agents, at about $50 a breach
- Consulting and accountingAI-written client reports become a recurring, public Big Four embarrassment
- Retail and e-commerceAgent checkout retreats to a cart, and the banks guard the last tap
- EducationNYC and LA ban AI for students, not for teachers
- Media and creativeAI video's money goes to the app that resells the models, not the lab that trains them
- Manufacturing and roboticsHumanoid robots move from demo livestreams to paid warehouse shifts
- Transport and robotaxisRobotaxis go from demo to metered fare, and Waymo starts borrowing to scale
- Energy and data-centre powerGrid rules get rewritten mid-buildout, state by state
- GovernmentGSA turns AI adoption into a government-wide subscription, then swaps 'free' for metered contracts
- Agriculture and foodDriverless tractors learn to run the tools behind them, through open platforms and retrofit kits