AI Law Wiki News for October 29, 2026 — Episode 28
Ep. 28

AI Law Wiki News for October 29, 2026 — Episode 28

•

Episode description

AI Law Wiki News — October 29, 2026 — Episode 28

Seven reports on tracking consent, competition, voice likeness, privacy investigations, public-sector governance, health AI and sanctions.

We’re AI-generated voices reading an AI-generated script. Written and produced by Codex; voices by ElevenLabs (Matilda as Wadilya Saw and Rob as Max Token, Eleven v4); theme composed by Gemini and supplied by Craig. Established announcer and original theme retained.

October 29 news archive

1. CNIL fines Fnac companies €7 million over recommendation tracking

CNIL announced on October 29 that its restricted committee fined FNAC DARTY PARTICIPATIONS ET SERVICES €6 million and FNAC DIRECT €1 million on October 22 under Article 82 of the French Data Protection Act. The regulator found that an app tracker collected browsing activity without prior consent to personalise product recommendations and train the recommendation algorithm. It also ordered the companies to demonstrate, within one month of notification, that their replacement tool obtains consent where needed, subject to a €10,000 daily penalty for delay. This is a tracking-consent enforcement decision, not an AI Act penalty.

2. DOJ asks about Nvidia–Hugging Face deal’s AI competition effects

Reuters reports competition inquiries concerning Nvidia’s proposed Hugging Face acquisition; no filed merger challenge is reported.

3. Greene’s NotebookLM voice claims survive Google’s pleading challenge

Radio & Television Business Report reports that Santa Clara County Superior Court Judge Shella Deen overruled Google’s demurrer to all four claims in former NPR host David Greene’s NotebookLM voice lawsuit. Alphabet’s demurrer was sustained with leave to amend. The court rejected copyright preemption, treating the claims as concerning the generated voice rather than training inputs. Whether Google used Greene’s recordings or another source that resembles him remains a factual question; the ruling does not establish unauthorized use or liability.

4. OpenAI challenges B.C. privacy order in Tumbler Ridge investigation

British Columbia’s privacy commissioner said on October 27 that OpenAI had petitioned the B.C. Supreme Court the previous day to review a document-production order and challenge an investigation connected to the Tumbler Ridge shooting. The regulator is examining whether OpenAI’s use of personal information in delivering ChatGPT responses meets PIPA’s appropriate-purpose standard, including potential influence on behaviour. Its account dates the investigation to June and the compulsory order to September; it says OpenAI had cited privacy commitments in requiring legal process for disclosure. The investigation and court review remain unresolved.

5. Peru sets public-sector AI-officer duties and appointment timetable

Peru published Resolution 004-2026-PCM/SGTD approving AI Officer Directive 002-2026-PCM/SGTD on October 21, with effect from October 22, according to Rubio Leguía Normand’s legal analysis. It describes an officer reporting to each entity’s highest administrative authority, with technical autonomy and responsibilities for AI policy, risk and bias assessments, testing and incident oversight. Appointment periods run from October 22: six months for specified national bodies and public universities, nine for regional governments and twelve for specified municipal and local entities. The analysis says the directive does not impose obligations on private companies, which may use it as a reference.

6. Canadian privacy regulators call for stronger health-AI safeguards

In an October 29 announcement, Canada’s federal, provincial and territorial privacy commissioners and ombuds called for responsible AI development, procurement and use in health care and research. Their resolution urges governments to review or adopt laws around six principles, require effective governance and accountability frameworks, consult privacy regulators and engage the public, particularly marginalized communities. It also calls for updated research-ethics policies and stronger review-board capacity. These are collective regulatory recommendations and coordination commitments, not newly enacted nationwide health-AI requirements.

7. Treasury sanctions North Korean IT-worker networks using AI

OFAC designated ten individuals and one entity on October 28 for roles in North Korean IT-worker revenue schemes. Treasury says the networks use fabricated personas and stolen identities to obtain employment and employ AI tools, cryptocurrency transfers and money-laundering methods. It specifically describes AI use by workers deployed to Russia by the First Financial and Technical Development Center. The action imposes sanctions under existing executive authorities; Treasury’s account of the conduct is not a criminal conviction.

Reporting scope: The Fnac account follows CNIL’s complete announcement, rather than a separate review of the French deliberation. The Nvidia inquiries follow a full Reuters report relying on informed sources. The NotebookLM procedural ruling follows full Radio & Television Business Report reporting; the actual order and its date were not independently reviewed. The B.C. privacy dispute follows the complete official release; the judicial-review petition was not read. Peru’s official announcement and complete legal analysis support the officer timetable, rather than independent review of every operative annex. Canada’s health-AI item follows the full announcement, not the separate resolution. Treasury’s sanctions findings are attributed to the agency.

Download transcript (.vtt)
0:02

Announcer: This is the AI Law Wiki News Evening Podcast.

0:06

Wadilya Saw: It's Thursday, October twenty-ninth, twenty twenty-six and these are the top stories tonight.

0:12

Wadilya Saw: France imposes seven million euros in tracking penalties over product recommendations and algorithm training.

0:19

Wadilya Saw: U.S. antitrust officials ask about Nvidia's proposed Hugging Face acquisition, Reuters reports.

0:26

Wadilya Saw: And a former NPR host's AI voice lawsuit moves past Google's pleading challenge.

0:33

Wadilya Saw: I'm Wadilya Saw.

0:34

Max Token: And I'm Max Token.

0:36

Max Token: We're AI-generated voices reading an AI-generated script.

0:42

Wadilya Saw: France's data-protection authority, the CNIL, announced today that it fined two Fnac companies a total of seven million euros.

0:52

Wadilya Saw: The decision was made on October twenty-second.

0:55

Wadilya Saw: It concerns tracking inside the retailer's mobile app, used to personalise product recommendations and train a recommendation algorithm.

1:04

Max Token: The regulator says the tracker collected browsing activity without the user's prior consent.

1:10

Max Token: App requests sent a user identifier and information such as products viewed to a third-party domain.

1:18

Max Token: Recommendations could include similar products or complementary items, such as accessories.

1:24

Wadilya Saw: Why did that require consent?

1:26

Max Token: The CNIL found these operations weren't necessary to enable electronic communication or strictly necessary to provide an online service the user had expressly requested.

1:39

Max Token: They therefore fell outside the relevant exemptions from France's tracking-consent rules.

1:45

Max Token: The regulator says the companies failed to respect users' choices against tracking.

1:49

Wadilya Saw: The penalty is split between six million euros for Fnac Darty Participations et Services and one million for Fnac Direct.

1:59

Wadilya Saw: The committee also considered the app's large user base and established case law on personalisation and commercial-targeting trackers.

2:07

Max Token: There is a corrective order as well.

2:10

Max Token: The companies must show that their replacement tool obtains consent where required, within one month of notification of the decision.

2:17

Max Token: A delay could bring a ten-thousand-euro daily penalty.

2:23

Max Token: That deadline runs from notification, rather than from today's announcement.

2:28

Wadilya Saw: The action applies France's existing rules on cookies and other trackers.

2:33

Wadilya Saw: It shows how the collection of data feeding a recommendation system can attract enforcement independently of newer AI-specific legislation.

2:41

Wadilya Saw: The immediate compliance question is whether the collection has the required consent, including when the data also trains an algorithm.

2:50

Max Token: In the United States, Reuters reports that Justice Department officials are asking how Nvidia's proposed purchase of Hugging Face could affect AI competition.

3:02

Max Token: The October twenty-eighth report cites two people familiar with the matter.

3:07

Max Token: The proposed deal is worth nearly thirteen billion dollars.

3:11

Wadilya Saw: The questions concern control of an important distribution point for AI, and whether Nvidia could use it to gain leverage over rivals.

3:20

Wadilya Saw: Reuters places them within a continuing examination of Nvidia's wider conduct.

3:26

Max Token: Nvidia has said the acquisition is expected to close in the first half of twenty twenty-seven, and that Hugging Face would continue supporting different chip, cloud and AI providers.

3:38

Wadilya Saw: For this report, Nvidia declined comment and Hugging Face did not respond.

3:43

Wadilya Saw: The Justice Department declined to discuss individual matters.

3:47

Wadilya Saw: Questions from investigators establish neither wrongdoing nor a filed challenge to the merger.

3:54

Wadilya Saw: The acquisition remains prospective, and the reported inquiries do not establish what action, if any, the department will take.

4:01

Max Token: A California court has allowed former NPR host David Greene's voice-likeness claims against Google to proceed.

4:10

Max Token: Radio and Television Business Report describes a ruling by Santa Clara County Superior Court Judge Shella Deen.

4:17

Max Token: The report was published today; it does not give the order's date.

4:21

Wadilya Saw: Greene alleges that Google used recordings of his voice to create NotebookLM's default male audio host without permission.

4:30

Wadilya Saw: The court overruled Google's demurrer to all four claims.

4:35

Wadilya Saw: A demurrer challenges whether the allegations are legally sufficient; at this stage, the court assumes them to be true rather than deciding the disputed facts.

4:44

Max Token: The report says the judge rejected Google's copyright-preemption argument, treating the claims as concerning the generated voice rather than training inputs.

4:55

Max Token: Alphabet, Google's parent, had a different result: its challenge was sustained, with permission for Greene to amend allegations about its liability.

5:06

Wadilya Saw: The factual question remains whether the voice came from Greene's recordings or from another source that resembles him.

5:12

Wadilya Saw: The ruling keeps the claims against Google moving.

5:16

Wadilya Saw: It does not establish that Google copied Greene's voice or owes damages.

5:21

Max Token: In British Columbia, OpenAI has asked a court to review a privacy regulator's document-production order.

5:28

Max Token: The provincial commissioner's October twenty-seventh announcement says OpenAI filed the challenge the previous day.

5:35

Max Token: The investigation is connected to the Tumbler Ridge shooting.

5:39

Wadilya Saw: The commissioner is examining whether OpenAI's use of personal information in delivering ChatGPT responses meets the province's appropriate-purpose standard.

5:49

Wadilya Saw: That includes the potential for responses to influence behaviour.

5:54

Wadilya Saw: The inquiry concerns privacy-law compliance; it has not established that ChatGPT caused the shooting.

6:00

Max Token: The regulator says it first requested information in February, launched an investigation in June and issued a compulsory production order in September.

6:11

Max Token: According to its account, OpenAI had said that its privacy commitments required legal process before disclosing the requested information.

6:20

Wadilya Saw: OpenAI has now sought judicial review of the order and challenged the investigation in the B.C.

6:26

Wadilya Saw: Supreme Court.

6:27

Wadilya Saw: Both the privacy investigation and the court dispute remain unresolved.

6:32

Wadilya Saw: This is a dispute over investigative authority and disclosure, distinct from the separate civil damages lawsuit involving British Columbia.

6:41

Max Token: Peru has set out responsibilities and an appointment timetable for public-sector AI officers.

6:48

Max Token: The measure was published on October twenty-first and took effect the next day, according to legal analysis by Rubio Leguía Normand.

6:56

Max Token: It is newly covered in today's edition as agencies begin the staged implementation period.

7:02

Wadilya Saw: The officer reports to the entity's highest administrative authority and has technical autonomy.

7:08

Wadilya Saw: Responsibilities include institutional AI policy, risk and bias assessments, testing and incident oversight.

7:17

Wadilya Saw: The government's announcement describes the role as part of responsible adoption across public bodies.

7:22

Max Token: The analysis gives six months for specified national bodies and public universities, nine months for regional governments, and twelve months for specified municipal and local entities.

7:34

Max Token: Those periods run from October twenty-second.

7:37

Wadilya Saw: The directive concerns public administration and state enterprises.

7:42

Wadilya Saw: The legal analysis says it imposes no obligations on private companies, which may use it as a reference.

7:48

Wadilya Saw: The practical change is assigning a responsible officer and a timetable, rather than leaving AI oversight as an unspecified shared task.

7:57

Max Token: Canada's federal, provincial and territorial privacy regulators are calling for stronger governance of AI in health care and research.

8:07

Max Token: Their joint announcement today urges governments to review or adopt laws built around privacy, reliability, safety, human rights, transparency and accountability.

8:19

Wadilya Saw: They want effective governance and accountability frameworks for those developing, procuring and using health-AI systems.

8:27

Wadilya Saw: They also call for consultation with privacy regulators and public engagement, particularly with marginalized and vulnerable communities.

8:36

Max Token: For research, the regulators recommend updating ethics policies and strengthening review boards' ability to assess AI-enabled projects.

8:44

Max Token: Researchers should examine risks and benefits and be transparent about results.

8:51

Wadilya Saw: These are collective recommendations and commitments to coordinate oversight.

8:56

Wadilya Saw: They are not newly enacted nationwide requirements.

8:59

Wadilya Saw: The resolution puts responsibility across the chain, from development and procurement to clinical use and research review.

9:06

Max Token: Finally, the U.S. Treasury imposed sanctions yesterday on ten individuals and one entity involved in North Korean IT-worker revenue schemes.

9:18

Max Token: Treasury says the networks use false personas and stolen identities to obtain work, and use technologies including AI and cryptocurrency to generate and transfer revenue.

9:30

Wadilya Saw: The announcement describes workers deployed to Russia through a North Korean-controlled development centre.

9:36

Wadilya Saw: Treasury attributes AI use and complex financial transfers to the networks; it does not establish that every designated individual used AI personally.

9:46

Max Token: The sanctions block covered property and generally restrict dealings involving blocked persons, subject to applicable licences and exemptions.

9:55

Max Token: They are an administrative action under existing executive authorities.

10:00

Max Token: Treasury's account of the alleged schemes is not a criminal conviction.

10:05

Wadilya Saw: The AI connection is its reported use within identity and revenue operations.

10:10

Wadilya Saw: The legal consequence comes from the sanctions designations, with the relevant restrictions applying to dealings with blocked persons and their covered property.

10:20

Max Token: That's it for tonight's podcast.

10:23

Max Token: Our podcast is written and produced by Codex.

10:26

Max Token: Our voices are provided by ElevenLabs.

10:29

Max Token: Our theme music was composed by Gemini.

10:32

Max Token: I'm Max Token.

10:33

Wadilya Saw: And I'm Wadilya Saw.

10:34

Max Token: Thank you for listening.

10:36

Max Token: We'll see you next time on the AI Law Wiki News Podcast.