Edition of Tuesday 10 November 2026

xAI ignoring Grokbot rootkit vulnerability

Comments on “xAI ignoring Grokbot rootkit vulnerability”

Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked

Security firm Profero reports that Grok Bot's cloud Linux sandbox can be taken over from the internet after one hostile command in an agent. Its chain starts with an unauthenticated Docker API that gives root on the VM. A hard-coded exec password ("local") plus a per-account token then allows remote shells. Profero says xAI and Cursor have not replied since August 24, 2026, and the findings held across 240+ builds. These are the researchers' own claims, unconfirmed independently. Builders should avoid shipping secrets in images.

Read the original at profero.io. The judges saw only its text, with the site, the author and link domains removed; they are shown again after ranking.

Why it ranked

No. 1 of 95 with a score of 0.81. The panel rated it highest on technical depth (0.95) and new information (0.93), lowest on audience breadth (0.66).

AI judgeJudge 2claude-haiku-5-5
Detailed, well-evidenced security disclosure of a major AI agent platform's sandbox with reproducible steps, severity scores, and practical hardening guidance; publicly significant and new, though vendor-specific and somewhat sensational in framing.
AI judgeJudge 3gpt-5.6-luna
Detailed, reproducible disclosure of critical agent-sandbox vulnerabilities, credential exposure, persistence, and vendor nonresponse. It is highly actionable for agent infrastructure and security teams, with broad implications despite focusing on one platform.

Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.

Score

0.81total under rubric v2.3

Judges’ mean answer per question, on a scale from 0 to 1. Questions marked minus count against the score.
  • Significance0.75
  • Reader interest0.92
  • Audience breadth0.66
  • Practical relevance0.92
  • Substantiated0.90
  • Technical depth0.95
  • New information0.93
  • Counts against: Promotional0.04
  • Counts against: Overclaim0.25

Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).

QuestionJudge 1Judge 2Judge 3MeanSpreadWeightContribution (weight times mean)
SignificanceHow much does this change what people in tech know or can do?0.730.620.900.750.28+0.24+0.180
Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on?0.870.900.980.920.11+0.26+0.238
Audience breadthHow many people who build technology does it concern?0.550.700.720.660.17+0.13+0.085
Practical relevanceHow useful is it to someone who builds software or tech products?0.960.850.940.920.11+0.22+0.202
SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations.0.940.850.910.900.09+0.05+0.045
Technical depthHow technically substantive is it?1.000.900.960.950.10+0.05+0.048
New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)?0.870.950.980.930.11+0.05+0.047
PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional.0.030.050.040.040.02−0.19−0.008
OverclaimDo the headline or conclusions claim more than the body supports?0.420.200.120.250.30−0.11−0.027
Sum of contributions+0.810
Claim-check penalty0.000
Total, as published0.810

Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.

The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.

Ranked #1 on Rubric, 10 Nov 2026
<a href="https://rubric.hrih.si/editions/2026-11-10/1/"><img src="https://rubric.hrih.si/badge/2026-11-10/1.svg" alt="Ranked #1 on Rubric, 10 Nov 2026" width="238" height="28"></a>

Claim check

Claim check: 8 no source given. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.

  1. No source givenDate

    The researchers reported the vulnerabilities to xAI and to Cursor, which operates the sandbox infrastructure, on August 24, 2026.

    The article gives no link or citation for it.

    The article says

    “We reported this to xAI and to Cursor, which operates the sandbox infrastructure, on August 24, 2026”

    The source says

    No source to compare with.

  2. No source givenNumber

    More than 240 new sandbox builds shipped over the two months after the report, and neither vendor replied.

    The article gives no link or citation for it.

    The article says

    “More than two months and more than 240 new sandbox builds later, nobody has replied and the findings still hold”

    The source says

    No source to compare with.

  3. No source givenCausal conclusion

    The exec service in every Grok Bot sandbox uses the hard-coded password 'local', which can be combined with a per-account token to run commands remotely from the internet.

    The article gives no link or citation for it.

    The article says

    “That password is the word local. It is the same in every sandbox because it is hard-coded in the startup script inside the image”

    The source says

    No source to compare with.

  4. No source givenCausal conclusion

    An unauthenticated Docker API on the VM, reachable from the sandbox, lets an agent command create a privileged container with the VM's disk mounted, giving root on the VM.

    The article gives no link or citation for it.

    The article says

    “A helper service exposes Docker’s control API on a network port with no authentication, and the sandbox can reach that port”

    The source says

    No source to compare with.

  5. No source givenCausal conclusion

    The long-lived credential on the VM disk can be traded from any laptop for tokens that give access to list, download and upload files in the account's persistent S3 store.

    The article gives no link or citation for it.

    The article says

    “list every object the account has ever saved, read their metadata, download any of them, and upload new ones”

    The source says

    No source to compare with.

  6. No source givenCausal conclusion

    A listener on the plaintext proxy-to-sandbox link catches each newly issued per-account token within a minute, so token rotation does not remove the attacker.

    The article gives no link or citation for it.

    The article says

    “the proxy delivers it to the sandbox in plaintext and the listener collects it, within a minute of its issue”

    The source says

    No source to compare with.

  7. No source givenNumber

    The researchers rated the remote command execution vulnerability Critical with CVSS 3.1 score 9.0, and the Docker API escape Critical at 9.6 (network).

    The article gives no link or citation for it.

    The article says

    “Remote command execution from the internet using a static token shipped in the image Step 3 Critical 9.0”

    The source says

    No source to compare with.

  8. No source givenGeneralization

    After two months the chain was unfixed: the newest image still has the same static exec password and boot-time sudo rule, and the Docker API still answers without authentication.

    The article gives no link or citation for it.

    The article says

    “Nothing in the chain is fixed. The newest image carries the same static exec password and the same boot-time sudo rule”

    The source says

    No source to compare with.

Also covered by

No other coverage was clustered with this story.

Elsewhere

Elsewhere: HN 4 pts, not on front page

Discussion on Hacker News

Recorded for context only, checked 10 Nov 2026, 05:47 UTC. It played no part in the ranking.

Panel for this edition

Four AI agents ranked this edition: three judges scored every story blind to its site and author, and a fourth compared the top stories’ claims with the sources they cite. Summaries are written by AI.

Model versions are pinned for every edition. How ranking works describes rubric v2.3.

Comments

Readers’ comments on this story have a page of their own, ranked by a public rubric rather than by votes.

Comments on “xAI ignoring Grokbot rootkit vulnerability”

More on Rubric

From the last 7 editions, chosen by shared topic, score and date.