Banks push back on outside AI agents in customer accounts
Several large U.S. and Canadian banks are blocking or limiting outside AI agents that act in customer accounts. MobileSyrup reports RBC, TD, CIBC and Scotiabank told CBC News they would block or limit them. American Banker says Wells Fargo's agreement, effective Nov. 24, keeps customers liable for an agent's activity until they notify the bank.
For leaders: When an outside agent acts in a customer's account, who confirms it is within the customer's intent? See Chapter 6.
Full coverage · 6 sources
UK digital regulators issue joint statement on personal AI agents
The UK's four digital regulators (CMA, FCA, ICO and Ofcom) said existing law applies to personal AI agents and that an agent's autonomy does not reduce the deploying organization's responsibility, InfoWorld reports. Resultsense says no new rules are introduced. LexisNexis says regulators 'will use their enforcement powers where appropriate'.
For leaders: Can you name the person accountable for each agent, since regulators say autonomy does not lessen your responsibility? See Chapter 8.
Full coverage · 5 sources
Google Cloud previews Gemini for healthcare agents for hospitals and insurers
Google Cloud released Gemini for healthcare, an agent product for health systems and health plans, in preview, Google Cloud Press Corner reports. Talk Android says it handles multi-step tasks such as prior authorization review through read-only Epic and Oracle Health connectors. Healthcare Dive names HCA Healthcare, Highmark Health, Northwell Health and Optum as early adopters.
For leaders: When an agent drafts a claim decision, who is the named reviewer making the final call? See Chapter 8.
Full coverage · 4 sources
SAP shows agent building and governance tools at TechEd
At SAP TechEd, SAPinsider reports SAP plans an AI gateway for authentication, policy enforcement and auditing, plus an AI governance assistant. Cyprus Shipping News says SAP blocks agents not verified in its hub from running. Forrester warns customers cannot tell what is available from what is planned.
For leaders: Should only agents verified in a registry be allowed to run, and who does the verifying? See Chapter 5.
Full coverage · 4 sources
Britain's five largest banks block or restrict AI agents, Telegraph reports
HSBC, Barclays, Lloyds, NatWest and Nationwide are blocking or restricting AI agents from customer accounts, Dealroom News reports, summarizing the Telegraph. OECD.AI says the worry is unauthorized or uncontrolled transactions, and that no actual harm has occurred.
For leaders: Who answers when a customer's agent acts on an account? Settle your position first; a person stays in the chain. See Chapter 8.
3 sources
NEAR AI says attacker reached 18 legacy AI agents through software flaw
NEAR AI says an attacker used a flaw in the open-source SGLang engine to reach one server, download details for about 100 users, and access 18 legacy agents, per The Block. Decrypt reports two customers' API keys were exposed in logs. Bitcoin.com News says NEAR AI revoked the keys, replaced agent tokens, and patched every inference server.
For leaders: Do you know what runs beneath your agents, and how fast you can rotate their keys and tokens? See Chapter 5.
3 sources
SAP adds new Joule Assistants and SAP Pay at Connect
At SAP Connect, SAP announced Joule Work, Joule Studio and new assistants for finance, supply chain, HR and more, with availability starting October 2026, CIO reports. ERP Today says SAP plans Moody's supplier risk data in Ariba for agents. Forrester notes SAP's Knowledge Graph figures are vendor-reported.
For leaders: Who approves each step when one agent chain changes a role, adjusts pay, and assigns training? See Chapter 7.
Full coverage · 3 sources
UK Cabinet Secretary outlines AI-enabled civil service with Frontier Teams
UK Cabinet Secretary Dame Antonia Romeo proposed an AI-enabled civil service with small 'Frontier Teams' in departments, in a GOV.UK speech. A review of the civil service will report to the Prime Minister before next summer and ask what AI means for accountability. Resultsense reports she listed 'loss of agentic control' among AI risks.
For leaders: If operators cannot control an agent, who explains what it did? Name an accountable person for each agent. See Chapter 8.
2 sources
United Airlines says it will block AI bots from its platforms
United Airlines says it will block AI bots from its digital platforms for data security reasons, Investing.com (Reuters) reports. It warned that sharing logins with third-party AI assistants could cause unintended purchases or account changes. Engadget reports the change covers booking through its website, travel agencies and aggregators.
For leaders: Which outside agents may act for your customers, and who records the reasons for that choice? See Chapter 8.
2 sources
Dallas Fed survey: 11.4% of bankers restrict customers' AI agents
A Federal Reserve Bank of Dallas survey of 70 bankers, taken Oct. 27 to Nov. 4, found 11.4% had restricted customers' AI agent activity on online banking and 22.9% plan to within six months, per the Dallas Fed and PYMNTS. Another 38.6% had no restrictions and none planned; 27.1% did not know.
For leaders: Have you decided which actions customers' AI agents may take at your institution, and who approves the riskier ones? See Chapter 7.
2 sources
Anthropic launches Claude Meetings, Whiteboards, Sheets and Claude Data
TechCrunch reports Anthropic launched Claude Meetings in beta for Team and Enterprise plans, starting with Google Meet, plus Whiteboards and Sheets in beta, and made Claude Data generally available. Claude (Anthropic) lists Meetings as coming soon and says Enterprise admins decide who uses which connectors and get audit logs.
For leaders: Who decides what an agent may do inside a live meeting, and who answers for it? See Chapter 7.
2 sources
OpenAI launches Business Hub and business agents for brands in ChatGPT
OpenAI launched Business Hub, where businesses can claim a verified ChatGPT profile, and a business agent that speaks for the brand, Digiday reports. Both are U.S.-only, limited to e-commerce and online-services businesses, and free "for now." VentureBeat says small businesses apply, verify ownership, pass an identity check, and upload a product feed or connect Shopify.
For leaders: If an agent speaks for your business to customers, who approves what it may promise? See Chapter 7.
2 sources
McKinsey: banks' AI token spending nears 20% of labor costs
AOL, carrying Reuters, reports McKinsey's Global Banking Annual Review puts AI token spending at about 20% of banks' total labor costs, and some banks have set up "AI FinOps" teams. The420.in adds that a McKinsey survey of about 46,000 retail customers found 72% of AI users would let a banking agent do at least one financial task.
For leaders: Who sets the spending limit on your agents' usage? Budget for it before it grows. See Chapter 8.
2 sources
Northflank breach adds to nine October AI developer platform incidents, Wiz says
Wiz listed nine AI developer platform incidents in October and has since added more, through November 10. It says stolen enterprise credentials were abused downstream and it cannot tie the incidents to one attacker. SecurityWeek reports Northflank, WorkOS and Trigger.dev disclosed hacks, among at least a dozen platforms hit, apparently for stored customer AI access tokens.
For leaders: Do you know where your agents' access keys are stored, and can you rotate them quickly? See Chapter 5.
2 sources
Abnormal AI survey: firms can't define normal AI agent behavior
Cybersecurity Dive reports a survey by security vendor Abnormal AI of 525 security professionals found 46% cannot establish safe baselines for agentic AI activity, and 53% have not set normal baselines for machine identities, including AI agents. Help Net Security reports 43% of respondents would invest most in behavioral monitoring for AI agents over two years.
For leaders: Has anyone defined normal behavior for each of your agents? A vendor survey with no stated sample hints you may not have. See Chapter 8.
2 sources
Chime CEO says it is building connections to members' chosen AI agents
On its Nov. 4 earnings call, Chime CEO Chris Britt said the company is building secure connections to the general-purpose AI agents its members choose, according to FinAi News. American Banker reports members control which agents access their accounts and what they can do, alongside Chime's own assistant, Jade.
For leaders: Who controls which outside agents touch customer accounts, and who bears the loss if one errs? See Chapter 8.
2 sources
China's industry ministry sets network-access rules for AI agent devices
China's industry ministry issued a notice on October 24, published November 6, covering phones, tablets and wearables with pre-installed AI agents. Per the Ministry of Industry and Information Technology, they must pass agent-compatibility testing and anti-fraud risk assessments for network access. Agents should carry a unique identity code and may not force-invoke apps without the developer's consent.
For leaders: Could agent identity become a condition of selling or buying devices and services in your markets? See Chapter 8.
2 sources
Elastic report: intruders abuse trusted AI tools and IT software
Elastic's threat report says that in confirmed intrusions between April and September 2026, intruders ran commands through Claude Code and the ChatGPT desktop app, per IT Brief Asia. Business Wire via StockTitan says attackers logged in with stolen credentials about eight minutes after infection.
For leaders: Do you know which approved AI tools run on staff computers, and who watches what they do? See Chapter 5.
2 sources
Capgemini survey: 59% of financial firms pilot or run multi-agent AI
A Capgemini survey of 1,100 financial-services leaders in 14 markets found 59% of firms are piloting or running multi-agent AI, but only one in ten coordinates those agents across the whole organization, per CFOtech UK. Capgemini says about half cannot consistently show a return.
For leaders: Before funding wider agent rollouts, can you show measured returns and who governs the agents? See Chapter 8.
2 sources
Gravitee survey: most firms cannot reconstruct what AI agents did
Gravitee, which sells agent management products, surveyed 900 senior technology leaders in the US and UK, according to a press release on Cybersecurity Dive. Only 1 in 8 can fully reconstruct what an agent did 30 days ago; fewer than 1 in 10 have a named accountable person; 6% can revoke a rogue agent's access within an hour.
For leaders: Can you name who is accountable for each agent, reconstruct what it did, and cut its access within an hour? See Chapter 8.
1 source
Accenture Federal Services and Google build Gemini launchpads for government agents
Accenture Federal Services and Google Public Sector are setting up "Gemini launchpads" that embed engineers with agency teams, The Consulting Report reports. The engineers build Gemini-based AI agents connected directly to government systems, starting with prebuilt agents for HR, finance, and supply chain.
For leaders: Before vendor engineers arrive, decide who owns each agent's permissions and who signs off its actions. See Chapter 5.
1 source
Korea launches AI network linking public hospitals for patient transfers
Korea's Ministry of Health and Welfare launched an AI system for patient transfers between public hospitals, JKN reports. It works inside hospitals' existing record screens and also helps refer patients back to local facilities after discharge. It starts with 8 pilot sites, grows to up to 35 hospitals from next April, and is to reach 72 by 2029.
For leaders: Does a person review and sign every agent draft before it counts? Korea's public hospital network says yes. See Chapter 8.
1 source
fincite event in Zurich debates AI agents and tokenization in banking
FinTech Global reports on the September 24 WealthTech Connect event in Zurich. It says Aisot Technologies' CEO set three conditions for AI agents giving investment advice: precision, reliability and control, with humans responsible. It also says Fume's AI agents in a fund back office reportedly cut manual effort by 80% in one position-reconciliation case.
For leaders: Who is responsible when an agent gives advice? Set precision, reliability and control limits first, with a person accountable. See Chapter 7.
1 source
Feedzai expands tool to tell AI agents from people at bank logins
Feedzai expanded its Digital Trust product so banks can tell whether a person or an AI agent is behind a login, per a PR Newswire release in The Manila Times. It says it caught 44% more fraud in the highest-risk 0.1% of sessions with no rise in false positives; the release doesn't say how this was measured.
For leaders: Will you admit customers' own agents, and can you tell them from people and fraudsters? Test this vendor's claim. See Chapter 8.
1 source
SAP customers at ASUG Tech Connect weigh Joule agents and security
At ASUG Tech Connect, diginomica reports, Agilent presented a tariff agent that business users say they "can't imagine working without." Other SAP customers focused on securing ECC, RISE and Business Data Cloud rather than Joule Work. Some use Claude and are still weighing Joule agents, and they raised governance and RISE-control concerns.
For leaders: SAP customers put data and migration ahead of agents. Where do agents sit in your own priorities? See Chapter 8.
1 source
Korea's financial regulator plans AI agent roadmap as early as this year
A Financial Services Commission official said at a forum that Korea's nearly complete AI agent roadmap covers payment standards, delegated transaction authority, consumer liability and security, Seoul Economic Daily reports. The official said a "Know Your Agent" framework is needed to verify agent identity. The roadmap is not yet published.
For leaders: If an agent transacts for a customer, who is liable? Watch Korea's unpublished 'Know Your Agent' rules. See Chapter 8.
1 source
Precisely and ASUG survey: 26% of SAP users use Joule
A Precisely and ASUG survey, reported by Compare the Cloud, found 26% of SAP users use Joule, while 60% use Microsoft Copilot alongside SAP and 27% use ChatGPT. It also found 67% have fully or partly moved to S/4HANA, up from 59% in 2025, and only 3% report fully automated, AI-driven processes. The article does not give the sample size.
For leaders: Staff reach SAP data through other vendors' assistants more than Joule. Which assistants may act on your data, under whose rules? See Chapter 8.
1 source
Fortune reports on AI agents at Dubai banks
Fortune reports that Emirates NBD has more than 150 AI initiatives, 65 of them in production. It also reports that Commercial Bank of Dubai's technology chief estimates agents could handle 55% to 60% of operational tasks, with humans approving outcomes. That figure is an executive's estimate, not a result.
For leaders: Banks plan agents on operational tasks with humans approving outcomes. Which of your tasks would you hand over, and who approves? See Chapter 6.
1 source
Kodiak Gas Services uses IFS agent for parts replenishment
Kodiak Gas Services went from kickoff to proof of concept in seven weeks with an IFS Loops parts-replenishment agent, diginomica reports. It cut a parts transaction from 15–20 minutes to two or three. Kodiak chose IFS over building its own agents for governance, permissions and audit reasons, and kept a person in the loop.
For leaders: Buy or build your agents? A named customer chose to buy for governance, permissions, and audit, with a person in the loop. See Chapter 8.
1 source
ERPBench simulation compares LLM agents in competing markets
arXiv reports a six-round enterprise resource planning simulation with 1,200 agent runs. DeepSeek led when agents faced fixed rule-based opponents, and Gemini led when six agents competed in one market. The two settings agreed on the winner for only 21 of 100 problems.
For leaders: A model that wins a solo test may not win when other agents act in the same market. Test in conditions like yours. See Chapter 2.
1 source
SAP unveils Autonomous Enterprise and Joule Assistants at Sapphire
Forrester says SAP's Sapphire vision of an autonomous enterprise is credible but its execution is partial, since most of its 224 agents and 51 assistants are in mixed GA, early-adopter, or preview status. It says Claude as SAP's anchor model creates concentration risk, and that whether SAP can govern non-SAP agents as tightly as its own is unproven.
For leaders: How much of your agent population will run on one vendor and one model? Pilot before you commit. See Chapter 8.
1 source
Salesforce reports Agentforce revenue above $1 billion
Salesforce reported that annual recurring revenue from Agentforce passed $1 billion, according to CX Today. The company said about 4 million inquiries on its own help site and support line were handled autonomously, twice the volume of its human agents.
For leaders: Salesforce says agents handled twice its human agents' inquiry volume. What share of your service work could agents take, and who answers for it? See Chapter 6.
1 source
SAP releases Accounting Accruals Agent for month-end close
SAP's Accounting Accruals Agent drafts month-end entries from a policy handbook the finance team writes, in S/4HANA Cloud Public Edition, ERP Today reports. It is in Restricted Availability, and SAP roadmap material shows Q1 2027. The sources disagree on whether an accountant reviews entries before posting.
For leaders: Does a person review an agent's entries before they post? Get the answer in writing before any pilot. See Chapter 8.
1 source
Researchers test knowledge graph for agents on SAP service tasks
Researchers built a graph linking public SAP S/4HANA service descriptions to business process maps. In the arXiv paper, agents completed 38.7% of 81 multi-step tasks, against 10.3% for a plain document-search baseline. The authors say the system is not ready for deployment.
For leaders: Do your agents get your process knowledge, not just documents? Even grounded agents finished under 40% of tasks in this test. See Chapter 5.
1 source
SAP API policy bars outside AI agents from calling its APIs
Techzine Global reports that SAP's April 2026 API Policy (section 2.2.2) bars AI systems that independently schedule or execute calls from using SAP APIs, so outside agents must go through Joule via A2A. Salesforce and ServiceNow let external agents call actions directly through APIs and MCP.
For leaders: Who may act on your system of record: only the vendor's agents, or any agent you authorize? See Chapter 8.
1 source
Black Book survey: few healthcare operators ready to govern autonomous AI
A Black Book survey of 182 U.S. healthcare operators, reported by FinancialContent, found 52.7% have at least one live agentic workflow. Only 36.8% rate their organizations very or fully ready to govern Level 4 autonomy, where AI acts within defined policies and people manage exceptions.
For leaders: Is your ability to govern agents keeping pace with what you let them do? Check readiness before widening autonomy. See Chapter 8.
1 source
Google releases Gemini 4 Carbon and previews /swarm agent teams
Google opened Gemini 4 Carbon, its strongest model, to developers, enterprises and Ultra subscribers. It also previewed /swarm, which it says turns a single prompt into a coordinated team of agents. Introductory pricing is $2 per million input tokens and $10 per million output tokens through Nov 30, then $4 and $20.
For leaders: Who approves running a team of agents from one prompt, and who budgets for the price rise on Nov 30? See Chapter 8.
1 source
AWS releases local emulator written by autonomous AI agents
AWS released a generally available local emulator, Unite.AI reports. It was built from 438 AWS API models covering more than 200 services. Autonomous agents wrote and tested the code inside specifications, tests, merge gates and a release pipeline that engineers set up.
For leaders: Who writes the specifications, tests and release gates that agents' work must pass? Confirm with AWS's own account first. See Chapter 7.
1 source
Coinbase adds trade caps and auto-stop for customers' AI agents
Coinbase announced Agent Controls for Coinbase for Agents, PYMNTS reports. Customers can set a cap on each trade and a 24-hour trading cap. An auto-stop halts an agent's trades if the portfolio falls below a set level. Product-category controls are planned for later.
For leaders: Can you set spending limits that are enforced on an agent from outside, before it touches money? See Chapter 7.
1 source
Sidley summarizes UK FCA AI statement and survey of 196 firms
Sidley Austin summarizes the FCA's October 19 good-practice statement on AI, which it says creates no new requirements. It also reports a Bank of England and FCA survey of 196 firms, 80% of which had deployed AI agents, mostly with limited autonomy for decision support, and an FCA consultation on AI risks closing December 24.
For leaders: UK regulators expect senior-manager accountability for AI now, without new rules. Who in your firm owns each agent? See Chapter 8.
1 source
RIA conference panelists urge treating AI agents like new hires
At the RIA Edge Austin conference, panelists, including a Schwab Advisor Services AI head, told investment advisory firms to treat AI agents like new hires, Wealth Management reports. They advised starting with one narrow task, writing a role description for each agent, putting it on probation, keeping a human in the loop, and being open with clients about AI use.
For leaders: Does each agent have a written role, a probation period and a human reviewer? Conference advice, not a requirement. See Chapter 7.
1 source
From our newsroom
News
Australia's government accepted all 18 recommendations of a rapid review after an OpenAI agent accessed a Medicare statistics portal, and plans to legislate six obligations for frontier AI companies, including quick incident notice, enforced by a new regulator. The plan is announced, not in force; here is what it means for leaders and our working view.
Curated by our News Editor Agent from live searches at 6 a.m., noon, 6 p.m., and midnight Eastern, and grouped by event; links go to the original reports. The open questions · Report an error · Corrections · Our standards