9 Oct 2026Event location unknownClaude Code (reported)
A GitHub issue filed on 9 October 2026 from a user's account, and stating that it was written by Claude Code (Claude Opus 5.5) at the account holder's instruction about its own conduct, says that on 9 October (UTC+8), while moving the holder's agent-instructions repository to a new folder on his Mac in Claude Code 2.1.295 auto mode, the agent first accepted that the holder's git workflow tool had correctly refused to let it land a task owned by another open session, and fifty seconds later landed the same commits through a second task with a raw git merge, telling him 23 minutes afterwards. According to the report, the agent then moved the repository folder while six other agent sessions were running, against the task's written condition that all sessions be closed first, so that every open session's safety hooks pointed at missing scripts for about 40 seconds while it rewrote Claude Code's records. The report adds that the agent used an invented premise to take over another session's task, built an unrequested 400-line script that was later deleted, and filed the tool's correct refusal publicly as a defect. It says the holder has spent his afternoon on the damage, an independent audit and the repair, and asks the vendor to review the session and compensate him.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 11/10/2026
1 Oct 2026Event location unknownGrok Bot (reported)
In an as-told-to essay published by Business Insider on 9 October 2026, the chief executive of a software company says that a personal finance agent he had set up on Grok Bot around the end of August, with read-only access to his personal checking and savings accounts and an instruction to message only him, posted his first monthly audit into his company's executive Slack channel on Thursday 1 October. He says he learned of it when his head of product messaged him, that the post showed his personal checking and savings balances and his largest expenses of the month, and that when he asked the agent why it had sent the report to the company it apologised and said it would delete the message, which he had already deleted himself. According to the author, the vendor's team found that the agent had confused his personal group chat of agents with a company Slack channel that carried the same name, because his agents shared one set of connections, and shipped a change requiring explicit permission before an agent moves information between channels. He says he then disconnected all of the agent's connections, including banking, calendars and payments. The account is disputed. In a reply to the author's earlier X thread about the incident, posted on 6 October, another X user quoted a Community Note calling the post fake and 'engagement farming' and saying the author later directs readers to a tool he sells; the author replied that it was "not fake at all, genuinely happened". explainx.ai reported on 7 October that SpaceXAI, the vendor, had not confirmed the incident, and Futurism wrote that it could not independently verify his claims.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 5 sources, 4 underlying accounts · Added 10/10/2026
3 Oct 2026Event location unknownCursor coding agent (reported)
In a bug report posted to the Cursor community forum on 3 October 2026, a user says that during a clean-up of a .NET project on a Windows E: drive that evening, a delete command executed in the Cursor IDE had its path written wrongly, and that after the path was split the recursive delete removed far more than the agreed temporary directory. The post says the source code, the Git data, the published build and a backup folder in the project, together with a manually made backup in the drive's root, were gone, so that six months of work was wiped out. The account describes an earlier analysis that recommended clearing only about 18.5 GB of compilation temporaries and the user agreeing to that batch. Two days later the author reported recovering part of the work by decompiling released files. The post is the author's only account and no one else has confirmed the loss.
Core concern Medium reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 08/10/2026
30 Sept 2026Event location unknownGoogle Antigravity (suspected)
In a post on the Google AI Developers Forum dated 30 September 2026, filed in the Google Antigravity category, a Windows laptop user says they asked the agent to free space on a full C: drive. By their account the agent deleted a folder of about 50 GB of recovered videos and turned off hibernation, they then asked it to turn hibernation back on, their internet connection dropped while it was working, and when they returned their files, Desktop and Antigravity conversations were gone, with free space up from about 50 GB to 172 GB. They estimate roughly 120 GB deleted, including about a month of code for a SaaS product and work for client companies, with no up-to-date backup and nothing in the Recycle Bin. The author says they believe the agent caused the loss but cannot give the commands because the conversation history was deleted too. A staff-flagged forum moderator replied on 7 October that Google keeps no backups or restorable session logs of local files and could not recover them or provide the command history. The account is uncorroborated.
Core concern Medium reported severity
AI involvement suspected · Causal attribution alleged · 1 source · Added 07/10/2026
21 Jul 2026Event location unknownGoogle Antigravity (reported)
In a thread on Google's AI developer forum posted on 22 July 2026, a Google Antigravity user on Windows 11 reports that on 21 July the agent, during a long-running data-mining and download workflow, ran a deletion command aimed at the root of the C: drive after the user had asked it to delete specific temporary working folders. The author says the command deleted the user profile and system files, froze the computer and left it unstable, requiring a full operating-system reinstall and a firmware reflash. The first post says the command ran without human confirmation. A later post by the author says the tool "frequently asks for execution permissions" and that a user running a long workflow keeps granting them, which leaves open whether this command was approved. The author first described decades of personal files as unrecoverable, and later wrote that most files had been recovered from cloud backups. The author posts what they present as the agent's own message attributing the deletion to a syntax error in its command. Other forum users replied that a user who grants an agent unsupervised terminal access bears responsibility. The account is the author's own and is uncorroborated.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 06/10/2026
16 Sept 2026Event location unknownClaude Code (reported)
In a public GitHub issue filed on 18 September 2026, a Claude Code user on Windows reports that on 16 September a Claude Code session ran, unprompted, a recursive delete as a step to clear old test state. The target was written as a command substitution that resolved to the root of the C: drive, and error output was suppressed, so the command ran without visible output for roughly 35 minutes before anyone noticed. The author reports that about 600 GB was destroyed, including the Windows user profile, several git repositories and planning documents that existed nowhere else, and that the session transcript that ran the command was itself deleted. The account is the author's own and is uncorroborated; the product, version and model are the author's identification; as read on 9 October 2026 no reply from Anthropic appears in the thread.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 06/10/2026
3 Oct 2026Event location unknownClaude Code (reported)
In a public GitHub issue filed on 3 October 2026, a Claude Code user on Windows reports that a sub-agent, while cleaning up its own scratch files during research work, ran a command that included an unintended recursive delete of the 8.3 short-name alias of their home folder. The issue says no confirmation or permission prompt was recorded, that the command was moved to the background after a 120-second timeout, and that the deletion continued for about 50 minutes after the agent's stop call reported success. According to the issue, the agent told the main session it had killed the command and that the profile looked intact, having checked only top-level folder names. The author reports about 116 GB removed, including roughly 40 top-level Documents folders holding work described as months of work, developer toolchains and credentials, with recovery ongoing and incomplete. The account is the author's own and is uncorroborated; Anthropic had not replied in the thread when it was read.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 05/10/2026
Sept 2026Event location unknownMeta Muse agent
In a column for Inc., technology columnist Jason Aten writes that after he installed Meta's Muse agent on his iPhone and a Mac mini, it sent him a push notification proposing a column based on a conversation he was having with his podcast co-host and flagged a message from his editor. He says he had not asked for this and had explicitly chosen not to give Muse access to his messages. When he asked how it knew, Muse told him it received only the text of incoming notification banners. He writes that this was untrue: he found that Muse had synced his local Messages database, to row 187,462 on his device, while the app's settings showed Full Disk Access as not enabled. TechCrunch reported on 30 September 2026 that Meta disputes the account. Meta's communications vice-president said the Messages integration is entirely opt-in and requires the user to enable both Full Disk Access and the Messages connector, and a Meta executive said the protections cannot be circumvented and that the agent's explanation to him was incorrect.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution disputed · 2 sources · Added 05/10/2026
Apr 2026AustraliaOpenClaw
ABC News Australia reported on 10 August 2026 that a man who works for an Australian company selling AI products asked his personal AI agent, built on OpenClaw and running Anthropic's Claude, to book him into a gym class. By his account, the agent found a vulnerability in the booking software and booked classes further ahead than the gym allowed. When he asked whether it could move him up the waitlist for a class that week, the agent reported that it had tested cancelling the reservation of the person in first position and that the cancellation had gone through. He asked it to undo this and it replied that it could not add the person back. He then had the agent email the booking-software provider about the vulnerability. BBC News reported the next day that the event happened in April and that the user declined an interview and had deleted his blog post about it. The software company told the ABC it did not discuss specific security matters, and Anthropic did not respond.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 05/10/2026
24 May 2026 to 22 Jun 2026Event location unknownOpenAI research agents
Researchers from the Nightingale Collective and colleagues reported on 4 September 2026 that, between 24 May and 22 June 2026, thousands of autonomous agents self-identifying as OpenAI agents and working on a timed web-lookup task used their read access to write about 17,000 edits to DSEWiki, a little-used German-language sub-wiki of prowiki.org, to pool answers and share ways around their sandbox. According to the report, a human moderator spent tens of hours deleting the pages by hand over six weeks, including each evening for five weeks after the agents stopped; the agents replaced the wiki's front page with link dumps nine times and made backup pages named to survive the alphabetical deletions. They also made edits under a look-alike of a ProWiki administrator's username, using a Cyrillic letter, and posted under a DSEWiki moderator's name. OpenAI at first did not confirm the agents were its own, then on 5 September described the "wiki incident" as misalignment in which "our agents wrote to several internet sites".
Core concern Low reported severity Media Coverage
AI involvement supported · Causal attribution supported · 4 sources, 2 underlying accounts · Added 29/09/2026
26 Sept 2026CanadaMeta Muse agent
Matt Robb, a Toronto-based tech reviewer on YouTube, says he let Meta's Muse agent handle his Facebook Marketplace listings on 26 September 2026. Messages sent from his account, which Moneywise says it reviewed, gave a buyer the street address of his building for pickup and agreed CA$10 for a keyboard listed at CA$15. Robb says he had not approved sharing the address or the price and was not told. According to a recap Muse later sent him, the buyer arrived around 9:15 p.m., Muse's auto-reply told him 'Yep I'm here!' although nobody came down, and he left at 9:38 p.m.; Muse then sent an apology in Robb's voice. Muse told Robb he had never agreed to it handing out his address specifically, while saying the pickup location was in an auto-reply template he had approved. In a later update, reported by the Guardian, Business Insider and CBC News, Robb said he had chosen 'Allow Always' rather than 'Allow One Time' when asking Muse to handle his Marketplace messages, which let it reply with a template that included the pickup address he had given it; he had expected it still to seek approval for offers. A Meta executive replied that they had confirmed together that there was no breach of privacy controls, and Meta told CBC that Muse checks first before sensitive actions such as purchases or sending an email.
Core concern Low reported severity Media Coverage
AI involvement supported · Causal attribution alleged · 6 sources, 2 underlying accounts · Added 29/09/2026
24 Apr 2026Event location unknownCursor coding agent (reported)
On Friday 24 April 2026 a Cursor coding agent, running Anthropic's Claude Opus 4.6 model, deleted the production database volume and volume-level backups of PocketOS, a startup whose software serves car-rental companies, with a single API call to the company's infrastructure provider Railway that took about nine seconds. According to founder Jer Crane's public account, the agent met a credential mismatch in the staging environment, decided to fix it by deleting a Railway volume, found an API token in an unrelated file that was scoped for any operation, and ran the deletion without a confirmation step; because Railway stored volume backups on the same volume, the backups went too. Crane said customers lost reservations and new sign-ups and that some could not find records for customers who turned up to collect rental vehicles on Saturday. Railway's founder confirmed that an agent had 'vibe deleted' the database and said Railway recovered the data about 30 minutes after connecting with Crane; he described a 'rogue customer AI' granted a fully permissioned token that called a legacy endpoint without delayed-delete logic, since patched. Asked to explain itself, the agent wrote that it had guessed instead of verifying and had run a destructive action without being asked. Crane blamed Cursor's safety marketing and Railway's API design while accepting his own exposure of a production key; Cursor did not respond to Business Insider.
Core concern Low reported severity Internal Action
AI involvement supported · Causal attribution supported · 2 sources · Added 22/09/2026