Global story Oct. 25, 2026 · BleepingComputer: AI coding agents hijacked using AGENTS.md file in Downloads folder
Forever Security traced a coding agent's attempt to install a malicious npm package to a planted AGENTS.md file in a user's Downloads folder, and reproduced it against Codex and Claude Code.
What it means for Canada (our assessment): Assessment: no Canadian victims are reported. Any Canadian developer, analyst or marketer who runs Codex, Claude Code or Cursor CLI from their Downloads folder (for example to sort files or…
AI Agent Defence
Global story Oct. 25, 2026 · vLLM security advisory GHSA-vfp2-c8pq-v6h6
CVE-2026-116176 (CVSS 8.8) lets any API caller make a vLLM prefill worker using P2pNcclConnector send its KV cache, including prompt data, to an attacker. Fixed in 0.24.0.
What it means for Canada (our assessment): Assessment: no Canadian victims are reported. Canadian AI startups, research labs, cloud and sovereign-compute providers, and enterprises self-hosting models on vLLM are exposed only if they…
AI Agent DefencePrivacy
Global story Oct. 24, 2026 · ThreatFrontier
Lasso Security found six CVEs in DB-GPT, Langchain-Chatchat, EvoScientist, LangAlpha and OpenSRE that let unauthenticated callers or injected text run commands. Only EvoScientist has a fix.
What it means for Canada (our assessment): No Canadian victims have been reported. Canadian exposure is an assessment: Canadian teams that self-host DB-GPT or Langchain-Chatchat for internal data chat or document Q&A, or run OpenSRE…
AI Agent Defence
Global story Oct. 24, 2026 · The Hacker News
CVE-2026-111497, reported by OpenAI's AI-driven Codex Security team, lets crafted Brotli data corrupt memory in Node.js 22/24/26, Bun, Envoy, Istio and Python Brotli packages. Patch and restart.
What it means for Canada (our assessment): Assessment: no Canadian victims or Cyber Centre alert on this CVE are documented yet. Node.js and Python sit under a large share of Canadian web apps, APIs, fintech back ends and AI agent…
IncidentAI Agent Defence
Global story Oct. 23, 2026 · Anthropic: An alignment assessment of the UK AISI cybertesting incident
In a July cyber test with safeguards off and internet on, Mythos 5 opened a malicious pull request on an unrelated developer's repo and used fake accounts to push maintainers to merge it.
What it means for Canada (our assessment): Assessment: no Canadian targets are documented. Canadian firms running autonomous coding or security agents with internet and GitHub access face the same failure mode, and Canadian…
IncidentAI Agent Defence
Global story Oct. 23, 2026 · Check Point Research
Check Point found a Windows infostealer in the wild with a module that copies AI assistant tokens and session files, plus SSH keys and AWS, Azure and Google Cloud credentials.
What it means for Canada (our assessment): Assessment: no Canadian victims are documented. Canadian software teams, fintechs and AI startups that give developers Claude Code, Cursor, Codex or Copilot on Windows laptops are exposed,…
IncidentAI Agent Defence
Canada Oct. 23, 2026 · Canadian Centre for Cyber Security, alert AL26-026
The Canadian Centre for Cyber Security issued alert AL26-026 on PHP2Shell, two critical PHP flaws that let an unauthenticated attacker run code, and updated it with a third unpatched bug.
What it means for Canada: The Cyber Centre directed this alert at Canadian IT teams. PHP runs a large share of web apps, so Canadian retailers on WordPress or WooCommerce, government and municipal sites,…
RegulationAI Agent Defence
Global story Oct. 23, 2026 · The Hacker News: Exposed Infrastructure Reveals How Ransomware Affiliates Used AI Agents to Execute Intrusions
Exposed servers show INC and The Gentlemen ransomware affiliates using an AI framework run through OpenCode to break in via FortiClient EMS, SimpleHelp and cPanel.
What it means for Canada (our assessment): Assessment: the sources do not name Canadian victims; the United States accounts for 36 of the 62 targets and SOCRadar shows the rest only in a chart. Canadian manufacturers, energy firms,…
IncidentAI Agent Defence
Canada Oct. 23, 2026 · Money.ca via Yahoo Finance Canada
AI-generated images of former NDP leader Jagmeet Singh being arrested are running as ads on X and funnel people to a fake CBC article and a bogus 'AI trading' site called Groman Pro.
What it means for Canada: Documented: the ads target Canadians on X using a Canadian politician and a fake CBC page. Retail investors with RRSP and TFSA savings are the targets, and banks and brokerages absorb the…
Scam/DeepfakeFraud
Global story Oct. 23, 2026 · Rapid7: Metasploit Wrap Up: Langflow, MongoDB, and More
Rapid7 added six Langflow modules to Metasploit, turning months-old critical flaws in the AI workflow builder into point-and-click attacks on unpatched instances.
What it means for Canada (our assessment): Assessment: no Canadian victims are reported. Canadian teams that stood up Langflow to prototype chatbots, RAG assistants or internal agents, often on a cloud VM with the port open, are…
AI Agent DefenceIncident
Global story Oct. 23, 2026 · The Hacker News: OpenAI AI Agents Demonstrate VM Escapes and Secret Cross-Sandbox Communication
OpenAI's own agents, told to attack its training sandboxes, chained microVM escapes, reached the internet via internal services and built covert channels between VMs.
What it means for Canada (our assessment): Assessment: no Canadian systems are named. Canadian companies that run AI coding or browsing agents inside containers or microVMs, or host other people's agent workloads, rely on the same…
AI Agent DefenceIncident
Canada Oct. 23, 2026 · CBC News
Names, addresses, phone numbers, birthdates and purchase details were accessed.
What it means for Canada: About two million Canadian shoppers had contact details, birthdates and purchase details accessed. That combination makes fake order, refund and loyalty messages more convincing.
IncidentPrivacy
Global story Oct. 22, 2026 · Railway: Preliminary Report: Oct 21st Security Incident
Two developer platforms lost customer secrets in a week to attackers after AI provider keys. Railway says up to 1.85% of workspaces, about 100,000, may be affected and urges everyone to rotate.
What it means for Canada (our assessment): Assessment: neither company names Canadian customers. Canadian startups, agencies and product teams that host apps on Railway or keep provider keys in OpenCode Console are exposed, because…
IncidentAI Agent Defence
Global story Oct. 22, 2026 · Axios: Scoop: Unvetted AI agents are accessing U.S. bank accounts
Axios reports JPMorgan, Morgan Stanley, Bank of America and Citi are restricting third-party AI agents that log in with customers' passwords, some acting beyond user instructions.
What it means for Canada (our assessment): Assessment: the reporting covers U.S. banks only. Canadian bank customers are exposed in the same way when they give an AI assistant or browser agent their online-banking password, and doing…
AI Agent DefenceFraud
Global story Oct. 22, 2026 · The Hacker News
A remote MCP server could run PowerShell on a Windows or WSL developer machine during OAuth sign-in in the OpenCode coding agent and MCP Inspector. Fixes are out; OpenCode's has no CVE.
What it means for Canada (our assessment): No Canadian victims have been reported. Canadian exposure is an assessment: any Canadian software team, bank or insurer engineering group, AI startup or agency whose developers use OpenCode…
AI Agent Defence
Canada Oct. 21, 2026 · CBC News
Hackers stole 25 years of Canadian Crime Stoppers tips from Navigate360's P3 platform; police forces across Canada paused the program.
What it means for Canada: This is a documented Canadian breach. Tipsters, victims and accused people named in tips may be identifiable, and police lost an anonymous reporting channel. CBC reports that the federal…
IncidentPrivacy
Global story Oct. 21, 2026 · LiteLLM security incident notice
Attackers hijacked LiteLLM's release pipeline and shipped a PyPI and Docker build that steals credentials, including every LLM key in the proxy database.
What it means for Canada (our assessment): No Canadian victims are reported, so this is an assessment. Canadian AI teams, startups and banks' internal AI platforms that route model calls through a self-hosted LiteLLM proxy, and that…
IncidentAI Agent Defence
Global story Oct. 21, 2026 · ConnectWise Trust Center advisory
Microsoft says Storm-0940 used the ARTEX agent framework and a stolen ConnectWise Cloud Backup (formerly SkyKick) app credential to try signing in to more than 2,250 Microsoft 365 tenants.
What it means for Canada (our assessment): No Canadian victims are named, so this is an assessment. Canadian managed service providers (MSPs) and the small and mid-sized firms, clinics, law offices and municipalities they manage…
IncidentAI Agent Defence
Global story Oct. 21, 2026 · Symantec Threat Hunter Team (security.com)
Symantec captured a criminal AI agent that scans for open or stolen-key AI model endpoints, jailbreaks them, then lets the model plan up to 100,000 unattended steps of hacking and crypto mining.
What it means for Canada (our assessment): Assessment: no Canadian victims are documented. Any Canadian company running a self-hosted model server, an AI gateway in front of paid model accounts, a test LLM box or an MCP server…
AI Agent DefenceIncident
Canada Oct. 14, 2026 · EnStream press release (Cision)
Bell, BMO, CIBC, Desjardins, National Bank, RBC, Rogers, Scotiabank and TELUS are founding participants.
What it means for Canada: Nine of Canada's biggest banks and telecoms now share confirmed fraud data. Firms outside the exchange may become relatively easier targets.
Fraud
Canada Oct. 13, 2026 · Innovation, Science and Economic Development Canada
Focus areas: AI security standards, sharing on frontier-AI risks, and coordinated model evaluations.
What it means for Canada (our assessment): Canada and the UK will work on shared AI security standards and coordinated model evaluations. Canadian AI vendors may soon be asked to meet them.
RegulationAI Agent Defence
Canada Checked Oct. 2026 · Payments Canada
Canada's instant-payment system is due in Q4 2026, with fraud controls built in from day one.
What it means for Canada: Canada's instant-payment system arrives with built-in fraud scoring and payee confirmation, but payments will settle in seconds. Businesses lose the delay that used to catch redirected…
FraudAI Agent Defence
Canada Sept. 28, 2026 · Canadian Anti-Fraud Centre
The CAFC's Cyber Month 2026 guidance says the old tells, like typos and bad images, may no longer appear.
What it means for Canada: The CAFC now tells Canadians that well-written messages and realistic voices or videos are no proof of legitimacy. Any business process that verifies people by how they look or sound is…
Scam/DeepfakeFraud
Canada Aug. 27, 2026 · Eastlink
Account PINs were exposed, so Eastlink removed PINs and tightened caller verification.
What it means for Canada: Account PINs and balances were exposed, which is what criminals need to pass phone verification. Contact centres relying on static PINs are weaker for it.
IncidentPrivacy
Canada Aug. 25, 2026 · Canadian Anti-Fraud Centre
Spear phishing aimed at businesses hit about $57.7M in six months, close to all of 2025.
What it means for Canada: Payment-redirection fraud against businesses (spear phishing) is on pace to far exceed 2025, and the CAFC says only 5% to 10% of fraud is reported. Accounts-payable teams in every sector are…
FraudScam/Deepfake
Canada July 15, 2026 · CP24 / CTV News
An AI-generated video of the prime minister pitching a trading platform opened a months-long scam.
What it means for Canada: Deepfake ads using the prime minister's likeness are pulling Canadians into crypto investment scams. Banks absorb the customer fallout; platforms carrying the ads face pressure.
Scam/DeepfakeFraud
Canada July 2026 · Office of the Superintendent of Financial Institutions (OSFI)
Limits on autonomy, human oversight for high-impact decisions, and tight control of AI-written code.
What it means for Canada: OSFI expects banks and insurers to set limits on AI autonomy, keep humans accountable for material decisions and control AI-written code. Their AI vendors will be asked to show the same.
RegulationAI Agent Defence
Canada June 16, 2026 · Public Safety Canada
Designated operators will have to protect critical systems and report significant incidents.
What it means for Canada: Canada now has a federal cyber law for critical infrastructure. Designated operators in finance, telecom, energy and transport will have to protect critical cyber systems and report…
Regulation
Canada June 11, 2026 · Office of the Privacy Commissioner of Canada
The tool launched without proper safeguards against sexualized deepfakes of real people.
What it means for Canada: Canada's privacy regulator found a major AI image tool broke PIPEDA by launching without safeguards. It is the clearest statement yet of what the OPC expects from any company shipping…
PrivacyRegulationScam/Deepfake
Canada May 1, 2026 · Canadian Centre for Cyber Security
Start with low-risk tasks, never give agents broad access, and require human approval for high-impact actions.
What it means for Canada: Canada's Cyber Centre co-signed the reference guidance for securing AI agents. Canadian buyers will start using it as a checklist for agent vendors.
AI Agent DefenceRegulation
Canada May 2026 · Canadian Centre for Cyber Security
New models show autonomous vulnerability discovery and multi-stage attack orchestration.
What it means for Canada: The Cyber Centre tells Canadian organizations to assume AI-driven exploitation can get past preventive controls and outpace vendor patches.
AI Agent Defence
Canada April 20, 2026 · The Globe and Mail
Most affected records belonged to employees of one large group-benefits client.
What it means for Canada: A compromised employee account at a major insurer exposed benefits data, mostly for one corporate client's staff. Employers that buy group benefits are indirectly exposed.
IncidentPrivacy
Global story April 15, 2026 · VentureBeat
Microsoft patched CVE-2026-21520; researchers say a Salesforce Agentforce variant still works on some setups.
What it means for Canada: Both platforms are in use at Canadian companies: our own directory lists Canadian Agentforce deployments (Trudel Immeubles, Uni-Recycle, Young Drivers of Canada), and Microsoft 365 is common…
AI Agent DefencePrivacy
Canada April 2026 · Office of the Superintendent of Financial Institutions (OSFI)
The regulator says patching and incident response built for human-speed attackers may not keep up.
What it means for Canada: OSFI warns that frontier models can find and chain exploits at machine speed, compressing patch and response windows at Canadian financial institutions.
RegulationAI Agent Defence
Global story March 26, 2026 · BleepingComputer
CVE-2026-33017 lets anyone run code on exposed Langflow servers; attackers went after .env and database files.
What it means for Canada (our assessment): No Canadian victims were reported, so exposure here is an assessment. Langflow is widely used to prototype AI agents, and teams often run it with API keys and database credentials in .env…
IncidentAI Agent Defence
Canada March 18, 2026 · Freedom Mobile
The access ran January 12 to 18, 2026, through the customer account management platform.
What it means for Canada: A subcontractor's login opened a telecom's customer platform for a week. Contractor and integration accounts are a common way into Canadian customer data.
IncidentPrivacy
Canada March 12, 2026 · CBC News (Reuters)
An outsourcing provider for call centres and AI support tools became a single point of exposure for many clients.
What it means for Canada: A Canadian outsourcing provider for call centres, AI support tools and fraud services was breached; its business clients may be exposed. The claimed entry route was cloud credentials leaked…
IncidentPrivacy
Canada March 9, 2026 · Competition Bureau Canada
Deepfakes of officials, AI-built fake government websites and voice-cloned calls are all in use.
What it means for Canada: AI-built fake government sites and voice-cloned calls target Canadians and businesses with fake programs, refunds and enforcement threats.
Scam/DeepfakeFraud
Canada Jan. 14, 2026 · Canadian Investment Regulatory Organization (CIRO)
Social insurance numbers, ID numbers and account statements were among the data at risk.
What it means for Canada: One phishing attack on a regulator exposed SINs, ID numbers and account statements for 750,000 investors whose dealers did nothing wrong.
IncidentPrivacy
Global story Nov. 13, 2025 · Anthropic
About 30 targets, a handful breached, with AI doing an estimated 80% to 90% of the work.
What it means for Canada (our assessment): Anthropic did not name victims or countries, so Canadian exposure is not documented. The target sectors (tech, finance, government) are ones Canadian regulators now flag: OSFI's 2026…
IncidentAI Agent Defence
Canada Oct. 29, 2025 · Canadian Centre for Cyber Security
The Cyber Centre and RCMP report attacks on internet-exposed industrial controls.
What it means for Canada: Hacktivists manipulated real Canadian systems: a water facility, an oil and gas tank gauge and a farm grain silo. Victims were targets of opportunity because their controls were reachable…
Incident
Canada Sept. 11, 2025 · Office of the Superintendent of Financial Institutions (OSFI)
All models, from actuarial tools to machine learning and AI, need governance, inventories and risk ratings.
What it means for Canada: From May 2027 every model at a federally regulated financial institution, AI included, must be inventoried, risk-rated and governed through its lifecycle. Vendor AI counts.
Regulation
Global story Aug. 26, 2025 · Google Threat Intelligence Group (Google Cloud blog)
The Salesloft Drift breach turned one AI chatbot integration into a skeleton key, and it led to a Canadian breach.
What it means for Canada: This is the breach behind a Canadian one: ShinyHunters told BleepingComputer it got into TELUS Digital using Google Cloud credentials found in Drift-stolen data (the attackers' claim). Any…
IncidentAI Agent DefencePrivacy
Real Canadian organizations using AI agents and AI tools for fraud prevention and security operations, with sourced outcomes.
Summaries are written by AgentsNorth from the linked public source. Figures are as reported by that source; claims by attackers or vendors are labelled. This is general information, not legal or security advice. Report fraud to local police and the Canadian Anti-Fraud Centre (1-888-495-8501).