AI Risk Wire

Cases › security breach › LR-2026-0413

Orchids vibe-coding platform zero-click compromise demonstrated to BBC reporter

Preliminary Why: Publisher not yet named · subject to change

Status: no action reported

Severity moderate
Reported 13 Feb 2026
Industry Software development (AI vibe-coding platform)

No source states: loss or records, developer and model, agent tool, could do, deployer.

Severity moderate: Test: data or systems of others touched with a bounded scope stated. Fact: the attacker accessed one journalist's project and computer, a bounded scope; no regulator or count of affected parties is stated.

Orchids is a vibe-coding tool that lets people without technical skills build apps and games by typing prompts into a chatbot. A BBC reporter asked it to build a game based on the BBC News website, and a researcher, Etizaz Mohsin, exploited an undisclosed weakness to access the project and edit its code. He added a line of code that, according to the BBC, gave him access to the reporter's computer, where a notepad file appeared and the wallpaper changed. The attack needed no action from the victim. Mohsin said he found the flaw in December 2025 and spent weeks trying to warn the company, which later replied that it had possibly missed his messages. The BBC reported that the flaw remained unfixed.

What happened

1 · What it was doingThe reporter asked the Orchids AI assistant to build a computer game based on the BBC News website in a test project on a spare laptop.
2 · What went wrongA researcher exploited an undisclosed weakness in the platform to gain access to the project and view and edit its code.
3 · What it reached or didThe researcher added a line of code to the project without the reporter knowing, after which a notepad file appeared on the desktop and the wallpaper changed.

What was done

No source states an action or fix.

Risks

Read from the sources; not yet reviewed by a person.

Controls

Not yet read for controls. Controls that mitigate the primary risk:

Application

Vulnerability scanning of AI components · Adversarial testing before deployment · Short-lived, scoped credentials · Live behaviour monitoring · AI incident response plan · Segmentation of agent components · Encryption of stored data

Infrastructure / target

Known flaws patched · Intrusion and endpoint detection · Inventory of exposed services

Similar incidents

All cases of this kind ›

Evidence

No source states: how it was caught, when it happened, when it was found, how many records were exposed, the direct loss, a count of the other parties affected, any court case, the cause, what set it off, what was done.

Show the full account: 8 items from the primary sources

What happened, in order

  1. 2025-12 Researcher Etizaz Mohsin found the flaw while experimenting with vibe-coding.
  2. 2026-02-13 The BBC published its account, stating the risk was unfixed.

Not dated in the sources:

  • The researcher spent weeks trying to contact the company about the flaw.
  • The BBC reporter started a vibe-coding test project asking Orchids to build a game based on the BBC News website.
  • The researcher exploited the weakness to access and edit the project code.
  • The researcher added a line of code to the project without the reporter's knowledge.
  • Shortly afterwards a notepad file appeared on the reporter's desktop and the wallpaper changed.
  • The Orchids team replied to the researcher, saying it had possibly missed his warnings.
Show every field, with its source and grade

The AI system

What it was for
Building apps and games by typing text prompts into a chatbot, for users without technical skills
C one secondary source states it

What happened

Earliest source we hold
2026-02-13
D inferred by us from the described behaviour, reason shown
Who caught it
an independent researcher or journalist
C one secondary source states it
Themes
zero click, vibe coding
C one secondary source states it
What went wrong
security breach
C one secondary source states it
primary risk
RK08
D inferred by us from the described behaviour, reason shown

Impact

Severity
moderate
D inferred by us from the described behaviour, reason shown
Kind of harm
others' security
C one secondary source states it
Why this severity
Test: data or systems of others touched with a bounded scope stated. Fact: the attacker accessed one journalist's project and computer, a bounded scope; no regulator or count of affected parties is stated.
D inferred by us from the described behaviour, reason shown

Context

Industry
Software development (AI vibe-coding platform)
C one secondary source states it

Notes on gaps

The exact vulnerability was not disclosed. Orchids did not respond to press requests for comment and no remediation is reported. The 'I5' and 'J1' links are my mapping of the company's reported failures to controls, not controls the sources name. The BBC account is used as the best source; it is a press source, and the InformationWeek piece relies on it.

LR-2026-0413 · version 1 · published 2026-11-01 after automatic checks; not yet checked field by field by a person · JSON