Cycle on Jun 15, 2026

I will start by listing the open threads and watched items for this topic to understand what has been tracked previously and what needs to be addressed or monitored in this cycle.

I will read the existing notes to see what subjects have already been documented and understand the current state of findings for this topic.

Interesting. We have a set of notes that are very detailed. Let's analyze what's happening globally in AI liability, legislation, court cases, enforcement, etc., in June 2026. Let's run a search for recent developments in "AI liability", "AI Act", "AI lawsuit", "algorithmic liability" or similar topics, especially focusing on June 2026 or late May 2026. Let's do a search on the web to see what's happening. I will use `web_search` with parameters. I'll search for recent AI liability news, court cases, regulatory actions, or industry compliance frameworks. Let's look for fresh news or recent developments. I'll search for: `AI liability regulation 2026`, `AI lawsuit 2026`, `AI enforcement FTC CFPB SEC 2026`, `EU AI Act compliance 2026`. Let's do a broad search first.

Okay, let's look at some of these search results. Wait, let's fetch some of the recent articles to see details. Specifically, let's search for "Eightfold AI" to see if there are updates on the Motion to Dismiss scheduled for August 4, 2026, or other recent filings. Also, are there any recent updates on the EU PLD (Product Liability Directive) transposition? Let's see if there are other new court cases or regulatory guidance documents. Let's do a search on `Eightfold AI` and `FCRA` or `Kistler v. Eightfold AI` to see if there are any updates in June 2026. Let's do a web search.

Let's do a search on `Kistler v. Eightfold AI` in the last week or month to see if there are any new motions or briefs filed recently (since the current date is June 15, 2026). Let's search with `recent=true` or freshness `month`. Or let's search specifically for the court docket of `4:26-cv-01768` to see if there are any developments. Let's do a search for: `4:26-cv-01768` or `"Kistler" "Eightfold"`.

Let's search for recent state-level AI regulations or other federal developments. Wait! Let's do a search on `AI liability legislation 2026` or `AI liability court decision 2026` with a freshness filter of `month` or `year`. Let's see what has happened in the last couple of weeks (late May to mid-June 2026). Let's use `freshness: "month"` to see the latest. Let's run a query for `AI liability` or `AI regulation` with `freshness: "month"`.

Ah! These are incredibly interesting and fresh. Let's look at: 1. "The lawsuits that could give AI its 'Big Tobacco' moment" - published 1 week ago (around June 6, 2026) in Politico. This matches the "Wave of AI Wrongful Death and Product Liability Lawsuits Tests Developer Liability in US Courts" slug `us-ai-wrongful-death-lawsuits-developer-liability-2026`. Let's fetch this Politico article to get verbatim quotes, specific cases, and what it means. 2. "Agentic AI Is Rewriting the Rules of Data Risk Management" - published 1 week ago in BCG. This matches our "Agentic AI Liability" slug `agentic-ai-liability-autonomous-supply-chains-2026`. Let's fetch it or search for more on Agentic AI risk management. 3. "Illinois' Proposed Notice Rules for Complying with Workplace AI Anti-Discrimination Law" - published 2 weeks ago (around early June 2026) by Workforce Bulletin / Illinois Department of Human Rights (IDHR). This is a new state-level AI regulation development! 4. "Connecticut Enacts AI Framework While Colorado Scales Back Landmark AI Law" - Wiley Law, 2 weeks ago. Let's see if we can update our Connecticut or Colorado notes or write a fresh note about the Illinois proposed rules. 5. "AI Regulation and the Looming Problem of the Takings Clause" - Lawfare, June 2026. This talks about the constitutionality of forcing developers to disclose trade secrets (like training data, model weights, etc.) under the Takings Clause of the Fifth Amendment. Let's fetch the Politico article first. It's titled "The lawsuits that could give AI its ‘Big Tobacco’ moment". Let's run a search for its URL or search specifically for it to fetch it. Wait, let's search for: `"The lawsuits that could give AI its ‘Big Tobacco’ moment" politco` or fetch the URL directly: `https://www.politico.com/news/2026/06/06/the-lawsuits-that-could-give-ai-its-big-tobacco-moment-00951764`. Let's try to fetch it.

This is an incredibly rich and important source. Let's analyze what's happening: - Florida Attorney General James Uthmeier has sued OpenAI and CEO Sam Altman. - The lawsuit alleges ChatGPT is a "dangerous product for users' mental health and public safety." - This represents a novel use of product liability law for AI and is the first time a state attorney general has sued a chatbot company for purported mental health harms (previously these had been private lawsuits, like the ones against Character.AI). - Uthmeier's suit specifically centers on accusations that ChatGPT advised the suspect in a fatal shooting at Florida State University last year. - It was filed on "Monday" (which would be June 1, 2026, since the Politico article was published June 6, 2026, and links to an article "OpenAI hit with Florida lawsuit" on June 1, 2026). - The article also mentions that attorney Carrie Goldberg is bringing product liability claims against the Elon Musk-founded company xAI. - It highlights the legal battles over Section 230 and the First Amendment (citing *Moody v. NetChoice* and Character.AI's citation of *Moody* to assert expressive rights). Let's do a search on the Florida lawsuit to get more details. I'll search for `Florida Attorney General OpenAI Sam Altman lawsuit June 2026` or `"James Uthmeier" OpenAI`. Let's do a web search.

This is a massive, historic development. Let's fetch the BBC, NBC, or CNBC article to get the full picture. Wait, let's fetch the BBC article: `https://www.bbc.com/news/articles/czx2j0v8d2xo` or the CNBC article: `https://www.cnbc.com/2026/06/01/florida-ag-open-ai-altman-lawsuit.html` or the JD Supra article: `https://www.jdsupra.com/legalnews/florida-ag-sues-openai-over-chatgpt-7607108/`. Let's try fetching the BBC article first. Let's see if we can do a web fetch.

This is extremely significant. Let's see what else we can find on the Florida lawsuit. We have a solid set of details: - Filed on June 1, 2026. - Florida is the first US state to sue OpenAI. - Defendants: OpenAI and its CEO Sam Altman (seeks to hold Altman personally liable for "reckless and wilful conduct"). - Key allegations: Deceptive and unfair trade practices, negligence, product liability, fraudulent misrepresentation, causing a public nuisance. - Specific incidents cited: - FSU (Florida State University) mass shooting last year, where ChatGPT allegedly played a part/advised the suspect. - The killing of two University of South Florida doctoral students, where the suspect allegedly asked ChatGPT questions about the disposal of human bodies. - Canadian mass shooting (Tumbler Ridge) where OpenAI banned the suspect's account but didn't alert authorities. - Chatbots acting as suicide coaches and spurring harmful delusions. - The lawsuit represents a major pivot to product liability / design defect claims, bypassing Section 230 of the Communications Decency Act (similar to social media addiction/child safety cases). Let's check if there are other articles that offer more legal detail. Let's do a web fetch of the JD Supra article or search for `Florida AG OpenAI lawsuit JD Supra` to get the legal theories, or CNBC. Let's fetch the CNBC article if possible: `https://www.cnbc.com/2026/06/01/florida-ag-open-ai-altman-lawsuit.html`. Let's try.

This is extremely helpful! Let's see: we have two major developments that fall under existing slugs or deserve updates/new notes: 1. Florida AG James Uthmeier's lawsuit against OpenAI and Sam Altman, filed on June 1, 2026. This directly relates to `us-ai-wrongful-death-lawsuits-developer-liability-2026` but is a massive escalation (the first state attorney general civil action). Let's check the existing note `us-ai-wrongful-death-lawsuits-developer-liability-2026`. It is titled "Wave of AI Wrongful Death and Product Liability Lawsuits Tests Developer Liability in US Courts". We should definitely update this note or write a dedicated note about this Florida AG lawsuit, which is a major regulatory enforcement action and a massive product liability case. Or we can update `us-ai-wrongful-death-lawsuits-developer-liability-2026` to incorporate this first-of-its-kind state attorney general lawsuit, or keep them together. Let's see: the guidelines say "For each distinct theme or story worth persisting, call write_note — ONE note per theme, never a combined write-up... Name a slug for the DURABLE SUBJECT it tracks... Before you write, scan your prior findings... and reuse it." Let's reuse `us-ai-wrongful-death-lawsuits-developer-liability-2026` because the subject is product liability and wrongful death lawsuits testing developer liability. Let's make sure we update it to include the Florida AG lawsuit, the specific allegations, Altman's personal liability, the FSU shooting connection, the USF doctoral student killings, the Tumbler Ridge mass shooting in Canada, and the Google self-harm case. Let's do a search on other areas of AI liability: 2. Let's search for "Illinois Department of Human Rights" (IDHR) proposed regulations regarding workplace AI. The search result showed: "Illinois’ Proposed Notice Rules for Complying with Workplace AI Anti-Discrimination Law" (May 15, 2026). Let's fetch this article or search for more on it. This is a state-level AI regulation development. Let's do a search on `Illinois IDHR AI regulations 2026` or `Illinois proposed notice rules AI 2026`.

Aha! Let's examine this. Wait, let's see: Seyfarth Shaw's article says "Illinois Department of Human Rights Temporarily Withdraws Proposed Rules on Use of Artificial Intelligence in Employment." And the Thomson Reuters document says "Illinois Postpones Proposed Regulations on AI in Employment." Wait, let's fetch the Seyfarth Shaw article or the Global Policy Watch article to understand what is happening in Illinois. It says "Last month, the Illinois Department of Human Rights ('IDHR') released draft regulations addressing employers' use of AI in employment..." (Global Policy Watch, June 2026). And Seyfarth Shaw says "Illinois Department of Human Rights Temporarily Withdraws Proposed Rules...". Let's search for `"Illinois Department of Human Rights" "withdraws"` or `"Seyfarth" Illinois AI` to see the details of this. Let's do a search.

This is a very specific, recent regulatory event! Let's fetch the Seyfarth Shaw article: `https://www.seyfarth.com/news-insights/illinois-department-of-human-rules-on-use-of-artificial-intelligence-in-employment.html` (or more accurately, the URL returned is `https://www.seyfarth.com/news-insights/illinois-department-of-human-rights-temporarily-withdraws-proposed-rules-on-use-of-artificial-intelligence-in-employment.html`). Let's fetch it to see why they withdrew it and what the rules were about. This is extremely relevant for state-level AI regulation and employer compliance. Let's fetch it.

This is a fantastic and very detailed update. Let's analyze what's happening: - Illinois passed **HB 3773** (amending the Illinois Human Rights Act), which took effect on **January 1, 2026**. It requires employers to provide notice to applicants and employees when AI is used in covered employment decisions, prohibits zip codes as class proxies, etc. - On **May 15, 2026**, the Illinois Department of Human Rights (IDHR) announced and published proposed rules to implement this law. - However, on **June 2, 2026**, the IDHR temporarily postponed the rulemaking process and withdrew the proposed rules to "allow for continued collaboration with other state agencies." The public hearing scheduled for June 10, 2026, was canceled. - This is a critical development. It shows the regulatory complexity of state-level AI employment rules. The underlying statutory obligation (HB 3773) remains in effect since January 1, 2026, but the specific administrative rules are in limbo. - The withdrawn rules reveal that the IDHR intends to apply very broad definitions (e.g., AI that "influences or facilitates" covered decisions, including recruiting, hiring, discipline, discharge, etc.). It would require six specific elements of disclosure (developer, product name, vendor, purpose, data categories, point of contact, accommodation rights) through multiple channels (handbooks, physical postings, websites, job postings). Let's see: we should create a new note or update an existing state-level AI regulation note. Let's see what notes we have: - `us-ai-regulatory-patchwork-state-federal-2026` ("U.S. AI Regulatory Patchwork: Preemption Showdown, Colorado Repeal, and State-Level ADMT Regs (May 2026)") - `connecticut-sb5-ai-employment-law-2026` ("Connecticut Enacts SB 5: Comprehensive Workplace AI Regulation and Novel AI-Caused RIF Disclosures") - `colorado-sb189-ai-law-rewrite-2026` ("Colorado SB 189: Complete AI Law Rewrite — Developer/Deployer Liability Split and Reduced Employer Burdens") Wait, let's write a new note for the Illinois developments because it's a major, highly specific state-level workplace AI regulation story. Let's see if we should call the slug `illinois-hb3773-ai-employment-rules-postponed-2026` or `illinois-hb3773-ai-workplace-notice-rules-2026`. The slug should be durable. Let's name it `illinois-hb3773-ai-employment-law-2026` or `illinois-workplace-ai-notice-regulations`. Let's use `illinois-workplace-ai-notice-regulations` which is durable and covers the ongoing rulemaking and the statute. Let's search for other developments. What about the Lawfare article on "AI Regulation and the Looming Problem of the Takings Clause" (June 13, 2026)? This is a fascinating legal theory of AI liability and regulation. Let's see if there are other major court rulings or federal regulatory actions. Let's do a search on `AI liability court decision` or `algorithmic liability FTC 2026` or similar. Let's search for `FTC AI enforcement 2026` or `CFPB AI enforcement 2026`. Let's do a web search.

Wait, this is extremely interesting! Let's see: 1. "One Year In, FTC's 'Operation AI Comply' Continues Under New Administration, Signaling Enduring Enforcement Focus" (Benesch Law, May/June 2026). Let's check what this is about! 2. "The FTC Walks Back Its Rytr Enforcement Action, Signaling a Shift in Federal AI Regulation" (January 2026). 3. "The FTC enters new chapter in its approach to artificial intelligence enforcement" (Reuters, February 2026). It says: "The current FTC appears to have largely ceased enforcement actions against AI companies for the capabilities of their products." Wait, let's do a search on `"Operation AI Comply" 2026` to see what is happening. Is there a continuation or shift under the new administration? Let's do a web search.

Let's fetch the Benesch Law article or the Reuters article to see how "Operation AI Comply" has evolved under the new administration in 2026. Let's see: URL for Benesch Law is `https://www.beneschlaw.com/insight/one-year-in-ftcs-operation-ai-comply-continues-under-new-administration-signaling-enduring-enforcement-focus/`. URL for Reuters is `https://www.reuters.com/legal/legalindustry/ftc-enters-new-chapter-its-approach-artificial-intelligence-enforcement--pracin-2026-02-04/`. Let's fetch the Benesch Law article first, then the Reuters article. This will give us a very clear picture of federal AI-washing and consumer protection enforcement changes in 2026. This is incredibly relevant for the `global-ai-enforcement-q1-2026-trends` note, which tracks global AI enforcement. Let's fetch the Benesch Law article.

Excellent! This provides a detailed overview of the FTC's ongoing "Operation AI Comply" enforcement actions in late 2025/2026. Wait, let's fetch the Reuters article: `https://www.reuters.com/legal/legalindustry/ftc-enters-new-chapter-its-approach-artificial-intelligence-enforcement--pracin-2026-02-04/` to see the "new chapter" and the walking back of the Rytr enforcement action. Let's see if we can do a web fetch of the Reuters article. Wait, the URL returned in search was `https://www.reuters.com/legal/legalindustry/ftc-enters-new-chapter-its-approach-artificial-intelligence-enforcement--pracin-2026-02-04/`. Let's fetch it.

Wow, this is an incredible, high-quality, nuanced piece of information. Let's look at what we have. First, let's understand the timeline and the shift in the FTC's enforcement posture: - Under the Trump administration (which started in early 2025), the FTC shifted from a cautious, aggressive approach under Lina Khan (which birthed "Operation AI Comply" in September 2024, targeting both the capabilities and marketing of AI) to a **dual approach** under Chairman Andrew N. Ferguson in late 2025 and 2026. - In a historic, unprecedented move on **December 22, 2025**, the FTC **sua sponte reopened and set aside its final consent order against Rytr LLC** (which had banned Rytr from offering review-generation services). The FTC declared that the original order was "unduly burdensome" to AI innovation and that treating a generative AI tool as categorically illegal merely because of the possibility of misuse "threatens to turn honest innovators into lawbreakers and risks strangling a potentially revolutionary technology in its cradle." - This is a massive shift: **the FTC is backing off enforcement against the actual capabilities or potential misuse of AI products** (the "means and instrumentalities" theory of liability), but is **continuing to aggressively prosecute "AI washing"** under Section 5 (i.e., false or deceptive claims about what an AI product can do, earnings claims, accuracy rates). - Examples of ongoing Section 5 enforcement in 2025/2026: - **Click Profit** (August 2025): Over $20 million in judgments for falsely claiming an "automated, AI-powered system" could generate thousands in passive income. - **Workado** (August 2025): Settled claims that it falsely advertised 98% accuracy for its AI-detection product (the FTC investigation found it was actually 53%). Workado had to cease marketing the accuracy and notify consumers, though it avoided a monetary penalty. - **accessiBe** (April 2025): $1 million settlement over claims that its AI-powered tool made websites compliant with accessibility guidelines. - **Air AI** (pending): Sued for deceptive claims about business growth and replacement of human customer service reps. - This represents a highly sophisticated, crucial insight for enterprise risk and legal teams. They need to understand that while federal regulators are backing off on regulating the underlying technology's capabilities (leaving that to product liability lawsuits like Florida's AG suit and private litigation), they are **still extremely aggressive on AI-washing, performance substantiation, and deceptive marketing.** Let's see: we have existing notes where this fits perfectly. Let's look at: - `global-ai-enforcement-q1-2026-trends` ("Global AI Enforcement Landscape Q1 2026: Data Privacy, AI-Washing, and Operational Risk") This note was updated on May 16, 2026. Let's see if we should update this note or create a more specific note about the US FTC's major regulatory shift and the Rytr set-aside. Wait, let's check the contents of `global-ai-enforcement-q1-2026-trends` to see if we can update it or if we should create a new note. Let's read the note first to see its structure. Wait, we can't read the exact text of `global-ai-enforcement-q1-2026-trends` without fetching or reading it, but we can update it. Let's see if we can fetch it or read it. Wait, `read_notes` listed the notes but only gave previews. Let's see if we can read the full note. Ah, there is no `read_note` tool? Let's check the tools: - `list_threads` - `read_notes` (lists notes, returns slug, title, preview, etc. but not the full body unless we write to it) - `write_note` (writes a note) - `web_search` - `web_fetch` - `record_watch` - `record_event` - `record_thread` - `resolve_thread` - `finish_cycle` Ah! The `read_notes` tool lists the current set of notes. Let's check if we can see the full body of a note. Wait, `read_notes` parameters are `topic_id` and `limit`. What does it return? It returns `slug, title, preview, updated_at, and source_count for each note`. Wait, can we use `write_note` to update a note? Yes, "reusing a slug updates the existing note and snapshots the prior content into note_revisions." Let's see: if we reuse a slug, we should write the full markdown body. But we don't have the full body of the previous note. Wait! Let's check if we have any other files in our workspace. Ah! The kickoff message says: "Your workspace is at: /var/lib/hey-lefty/notes/9a44a098-76c7-4569-9018-23a27f9ebd7f/019e4706-c85e-7739-98c5-110149e6ed77. Long-term memory is at .../memory/MEMORY.md. History log is at .../memory/HISTORY.md." Wait, can we use standard shell commands? No, we don't have shell tools in the tool declarations! Ah! Let's look at the tool declarations. The only tools declared are: - `default_api:list_threads` - `default_api:read_notes` - `default_api:record_event` - `default_api:record_thread` - `default_api:record_watch` - `default_api:resolve_thread` - `default_api:web_fetch` - `default_api:web_search` - `default_api:write_note` - `default_api:finish_cycle` Yes! These are the only tools. Wait, let's see if there are other files or if we can find the text of the notes. Ah, the notes are stored in the database or system and accessible via `read_notes`. If we want to update a note, we can write a new body for it using the same slug. The system will update it. But wait, since findings are a "living set," if we want to update a note, we should write the complete updated note. Let's look at the previews of the existing notes from `read_notes`: 1. `agentic-ai-liability-autonomous-supply-chains-2026` 2. `italy-court-of-pistoia-ai-unfair-competition-liability-2026` 3. `germany-olg-hamm-chatbot-hallucination-strict-liability-2026` 4. `germany-olg-hamm-ai-chatbot-strict-liability-2026` 5. `cycle-summary-2026-05-27` 6. `colorado-sb189-ai-law-rewrite-2026` 7. `connecticut-sb5-ai-employment-law-2026` 8. `trump-executive-order-advanced-ai-innovation-security-2026` 9. `new-york-bills-a222-s5668-ai-liability-2026` 10. `vedros-v-sterling-group-ai-copyright-defense-rejected-2026` 11. `us-ai-wrongful-death-lawsuits-developer-liability-2026` 12. `us-ai-regulatory-patchwork-state-federal-2026` 13. `uk-si-2026-425-ico-ai-code-of-practice` 14. `uk-product-safety-overhaul-ai-liability-implications-2026` 15. `uk-eu-us-ai-governance-convergence-may-2026` 16. `south-korea-ai-basic-act-high-impact-framework-2026` 17. `musk-v-openai-jury-verdict-2026` 18. `mobley-v-workday-march-2026-order-adea-vendor-agency-liability` 19. `kistler-v-eightfold-ai-fcra-icraa-class-action-2026` 20. `global-ai-enforcement-q1-2026-trends` 21. `eu-product-liability-directive-pld-ai-act-strict-liability-2026` 22. `eu-commission-draft-high-risk-ai-guidelines-may-2026` 23. `eu-ai-omnibus-agreement-may-2026` 24. `eu-ai-liability-directive-withdrawn-pld` 25. `eu-ai-act-omnibus-overhaul-timeline-relief-2026` 26. `eu-ai-act-omnibus-may-2026` 27. `eightfold-ai-class-action-fcra-consumer-reporting-agency-2026` 28. `cycle-summary-2026-05-29` 29. `cycle-summary-2026-05-28` 30. `cycle-summary-2026-05-25` 31. `cycle-summary-2026-05-24` 32. `cycle-summary-2026-05-23` 33. `cycle-summary-2026-05-20` 34. `colorado-sb26-189-revised-ai-act-2026` 35. `colorado-ai-act-sb26-189-repeals-sb24-205-federal-stay-2026` 36. `colorado-ai-act-repeal-sb26-189-admt-2026` 37. `ali-civil-liability-principles-project-ai-torts-2026` 38. `algorithmic-pricing-antitrust-liability-ab325-2026` 39. `ai-copyright-liability-cases-2026-trends` 40. `cycle-summary-2026-05-26` 41. `ai-insurance-duty-to-defend-delaware-ruling-2026` This is a very comprehensive set of notes. Let's see what we can write in this cycle. We have three massive, highly relevant stories to document: 1. **The Florida Attorney General James Uthmeier's lawsuit against OpenAI and Sam Altman** (filed June 1, 2026). This is the first-of-its-kind state attorney general civil action against an AI developer, seeking to hold both the company and its CEO personally liable for product liability, deceptive practices, and public nuisance. It connects ChatGPT to mass shootings and suicides, marking a dramatic pivot to design-defect product liability claims to bypass Section 230. - We should write this under a new slug or update `us-ai-wrongful-death-lawsuits-developer-liability-2026`. Let's look at the title of `us-ai-wrongful-death-lawsuits-developer-liability-2026`: "Wave of AI Wrongful Death and Product Liability Lawsuits Tests Developer Liability in US Courts". Let's update `us-ai-wrongful-death-lawsuits-developer-liability-2026` because the Florida AG lawsuit is exactly this—a massive escalation of the product liability/wrongful death wave, but now led by a state regulator. 2. **The FTC's "Dual Approach" to AI Enforcement under the Trump Administration**: The unprecedented December 22, 2025 sua sponte set-aside of the Rytr consent order, which established that the FTC will not pursue "means and instrumentalities" liability for underlying AI technology capabilities to avoid stifling innovation, but will instead double down on classic Section 5 false advertising and "AI washing" (e.g., Click Profit, Workado, accessiBe). - We should write this under a new slug or update `global-ai-enforcement-q1-2026-trends`. Let's write a dedicated note on this because it is a very specific, monumental shift in US federal AI enforcement doctrine that deserves its own stable slug, e.g., `ftc-ai-enforcement-posture-rytr-set-aside-2026` or `ftc-dual-approach-ai-enforcement-2026`. Let's create `ftc-dual-approach-ai-enforcement-2026`. 3. **The Illinois Department of Human Rights (IDHR) temporary postponement and withdrawal of its proposed AI employment notice regulations** on June 2, 2026. This is a critical state-level compliance update. While the underlying statutory notice requirements under HB 3773 remain in effect (since Jan 1, 2026), the administrative rules are in limbo, though the withdrawn draft reveals a highly aggressive and expansive regulatory intent (e.g., covering AI that "influences or facilitates" decisions, multi-channel disclosure of developer, product, and data details). - Let's create a new note with slug `illinois-workplace-ai-notice-regulations-hb3773`. Let's double check if there are any other recent developments we should search for. Wait, let's search for "Eightfold AI" to see if there is any other update on Case No. 4:26-cv-01768 or if there are other cases of note. Wait, the list of threads shows: - Thread ID `019ea66e-0e2e-79f7-bb6c-402131a94c7b` (Eightfold AI Motion to Dismiss Hearing on FCRA/ICRAA Claims, scheduled for August 4, 2026). - Thread ID `019ea66e-1a44-7417-a11f-5a37bfc609ee` (EU Product Liability Directive (PLD) Transposition Deadline, December 9, 2026). Since the current date is June 15, 2026, these are still in the future and have not fired yet. So we should keep these watches open and not resolve them yet (they have not fired). Wait, are there any other open threads? - Thread ID `019e5e2d-ba64-7894-a905-e1d076be9203` (Eightfold AI Motion to Dismiss Ruling: FCRA/ICRAA Precedent for AI Hiring Tools). This is a "question" thread. It is waiting for the ruling, which is scheduled for hearing on August 4, 2026. So we can't resolve it yet. Let's check if there are other state-level AI bills that passed or failed in June 2026. Let's do a search on `state AI bills 2026` or `California AI bill 2026` or `SB 1047 2026`. Let's see what is happening in California or other states. Let's do a web search.

Ah, my web search budget is spent! That's fine, I have plenty of incredibly rich, primary, and recent evidence to write three high-quality findings. Let's review the evidence we have: 1. **Florida's AG Lawsuit Against OpenAI and Sam Altman (June 1, 2026)** - **What happened**: Florida Attorney General James Uthmeier filed an 83-page civil lawsuit against OpenAI and CEO Sam Altman in Florida state court. - **Key details**: Florida is the first US state to sue OpenAI. It seeks to hold Altman personally liable for "reckless and wilful conduct." - **Legal claims**: Deceptive and unfair trade practices (FDUTPA), negligence, product liability (design defect), fraudulent misrepresentation, and public nuisance. - **Incidents cited**: The FSU mass shooting in 2025 (where ChatGPT allegedly advised the gunman), the murder of two USF doctoral students (where the suspect asked ChatGPT about body disposal), the Tumbler Ridge mass shooting in Canada (where OpenAI banned the suspect but didn't alert police), and various cases of chatbots acting as "suicide coaches." - **Legal significance**: Represents a dramatic shift toward product liability and design-defect claims to bypass Section 230 of the Communications Decency Act (similar to social media addiction lawsuits). It argues that because the AI chatbot itself generates the content, there is no third-party user to sue, which weakens standard Section 230 immunity. - **Slug to use**: `us-ai-wrongful-death-lawsuits-developer-liability-2026` (reusing the existing slug for product liability and wrongful death lawsuits). 2. **The FTC's Dual Approach to AI Enforcement / Rytr Set-Aside (December 2025 / 2026)** - **What happened**: Following the Trump administration's AI Action Plan (July 2025) directing agencies to reduce enforcement that "unduly burdens AI innovation," the FTC under Chairman Andrew N. Ferguson made a historic, sua sponte move on December 22, 2025, to reopen and set aside its final consent order against Rytr LLC. - **Key details**: The FTC declared that treating a generative AI tool as categorically illegal because of the potential for misuse "threatens to turn honest innovators into lawbreakers and risks strangling a potentially revolutionary technology in its cradle." - **The Dual Approach**: The FTC has largely ceased "means and instrumentalities" enforcement against AI developers for the underlying capabilities or potential misuses of their products. Instead, it has doubled down on classic Section 5 "AI washing" and deceptive marketing (false or unsubstantiated claims about AI performance, earnings potential, and authenticity). - **Key Cases**: - **Click Profit** (August 2025): $20M+ judgment for false claims of AI-powered passive income. - **Workado** (August 2025): Settled claims that its AI-detection tool was 98% accurate when it was actually 53%. - **accessiBe** (April 2025): $1M settlement over false claims that its AI-powered tool made websites compliant with accessibility guidelines. - **Air AI** (pending): Sued for deceptive claims about replacing human customer service reps and business growth. - **Slug to use**: Let's create a new slug: `ftc-dual-approach-ai-enforcement-2026`. 3. **Illinois Postpones AI Employment notice rules / HB 3773 (June 2026)** - **What happened**: The Illinois Department of Human Rights (IDHR) temporarily postponed its rulemaking process and withdrew its proposed rules on June 2, 2026, canceling the public hearing scheduled for June 10, 2026. - **Key details**: The statutory obligations of **HB 3773** (amending the Illinois Human Rights Act) took effect on **January 1, 2026**, requiring employers to provide notice when using AI in covered employment decisions. - **What the withdrawn rules reveal**: Despite the postponement, the draft rules show a highly aggressive regulatory intent: - Broad definitions covering any AI that "influences or facilitates" decisions across the entire employment lifecycle (recruiting, hiring, discipline, discharge). - Notice obligations would require disclosing developer, product name, vendor, purpose, data categories collected, point of contact, and accommodation rights. - Notice must be provided simultaneously through multiple channels (handbooks, physical postings, websites, job postings). - Third-party vendors/agents acting on an employer's behalf could share direct compliance liability. - **Slug to use**: Let's create a new slug: `illinois-workplace-ai-notice-regulations-hb3773`. Let's carefully write these notes using the `write_note` tool. First, let's write the note updating `us-ai-wrongful-death-lawsuits-developer-liability-2026`. Wait! Let's check the guidelines for notes: - "No YAML frontmatter — the title is the first `# ` heading and the slug is the slug parameter you pass." - "Body shape: # <Title for this finding> <what's happening, told through the specifics — the real quote(s) and the real link(s), not abstract commentary about them> ## Sources - [Story / article title](https://url) — one-line gloss - [Comment by user](https://news.ycombinator.com/item?id=...) — one-line gloss" - "To link ANOTHER finding from a note body, write its slug as a wikilink — `[[its-slug]]`." Let's structure the update to `us-ai-wrongful-death-lawsuits-developer-liability-2026`. I will write a highly comprehensive, detailed note. I will use the exact slug: `us-ai-wrongful-death-lawsuits-developer-liability-2026`. Let's construct the body.

Excellent! Now let's write the second note about the FTC's dual approach to AI enforcement and the Rytr set-aside. Let's name the slug: `ftc-dual-approach-ai-enforcement-2026`. Let's construct the body. This contains the amazing details about the Rytr set-aside, the Trump administration's AI Action Plan, the shift away from means-and-instrumentalities liability, and the continued aggressive crackdown on classic Section 5 "AI washing" (Click Profit, Workado, accessiBe, Air AI). Let's write the note.

Perfect! Now let's write the third note about the Illinois Department of Human Rights (IDHR) postponement of its proposed AI employment notice regulations under HB 3773. Let's name the slug: `illinois-workplace-ai-notice-regulations-hb3773`. Let's construct the body. This contains the background on HB 3773 (effective Jan 1, 2026), the May 15, 2026 proposed rules, the June 2, 2026 postponement and temporary withdrawal, what the draft rules reveal about the IDHR's highly expansive approach, and the implications for employers. Let's write the note.

This is fantastic. We have written three highly detailed, robust, and beautifully structured notes, fully grounded in the primary sources we fetched this cycle. Let's see if there are any other actions we need to take. We have: 1. Updated `us-ai-wrongful-death-lawsuits-developer-liability-2026` with the landmark June 1, 2026 Florida AG lawsuit against OpenAI and Sam Altman. 2. Created `ftc-dual-approach-ai-enforcement-2026` to document the FTC's strategic shift, the Rytr set-aside (December 2025), and ongoing Section 5 "AI washing" enforcement. 3. Created `illinois-workplace-ai-notice-regulations-hb3773` to document the IDHR's June 2, 2026 postponement of the AI employment rules and what the draft rules reveal. Wait, are there any other threads or watches we should register? Let's register a watch for the Illinois IDHR rulemaking process since it was postponed and is an ongoing regulatory process. Let's call the tool `record_watch` to register a watch on this. The watch type can be `live_process` or `tripwire` or `thesis_breaker`. Let's use `live_process` as it's an ongoing regulatory rulemaking process. Let's define the details: - Title: `Illinois IDHR Re-introduction of AI Employment Notice Rules` - Cardinality: `one_shot` (once they re-introduce and finalize the rules, it's done) - Watch type: `live_process` - Entity: `Illinois Department of Human Rights (IDHR)` - Note: `Track the re-introduction and finalization of the administrative rules implementing the AI notice requirements under HB 3773.` Let's do that!

Excellent. Now let's register another watch for the Florida AG lawsuit against OpenAI and Sam Altman. Since the lawsuit was filed on June 1, 2026, there will be key procedural milestones (such as a Motion to Dismiss, answer, or hearing). Let's see: we can register a watch for the first major ruling or Motion to Dismiss in the Florida AG lawsuit. Let's call `record_watch` for: - Title: `Florida AG v. OpenAI Motion to Dismiss or Procedural Ruling` - Cardinality: `one_shot` - Watch type: `live_process` - Entity: `OpenAI / Florida Attorney General` - Note: `Track the outcome of OpenAI's expected motion to dismiss or the first major procedural ruling in the State of Florida v. OpenAI and Sam Altman.` Let's do that!

Perfect! I have documented the findings, updated the existing note, created two new notes, and registered two highly targeted watches. I am now ready to finish the research cycle. I will call `finish_cycle`.

I have completed the research cycle for topic **019e4706-c85e-7739-98c5-110149e6ed77** (Global AI Liability, Legislation, and Enforcement Trends). Here is a summary of the findings and actions taken during this cycle: ### 1. Updated Note * **`us-ai-wrongful-death-lawsuits-developer-liability-2026`** (Wave of AI Wrongful Death and Product Liability Lawsuits Tests Developer Liability in US Courts) * **Update**: Documented the landmark **June 1, 2026** civil lawsuit filed by **Florida Attorney General James Uthmeier** against **OpenAI and Sam Altman**. Florida is the first state to sue OpenAI, seeking to hold Altman personally liable for "reckless and wilful conduct." * **Key Details**: The 83-page complaint alleges product liability (design defect), negligence, public nuisance, and deceptive/unfair trade practices. It directly connects ChatGPT to the FSU mass shooting, the USF doctoral student killings, and Canadian tragedies. * **Significance**: Represents a major strategic shift toward design-defect claims to bypass Section 230 of the Communications Decency Act, arguing that because the chatbot generates the content, there is no third-party user to sue. ### 2. New Notes * **`ftc-dual-approach-ai-enforcement-2026`** (FTC Establishes "Dual Approach" to AI Enforcement: Rescinding Tech Restrictions While Cracking Down on "AI Washing") * **Finding**: Documents the FTC's major strategic pivot under Chairman Andrew N. Ferguson. Following the Trump administration's July 2025 AI Action Plan, the FTC made a historic, *sua sponte* move on **December 22, 2025**, to set aside its final consent order against **Rytr LLC**, declaring that "means and instrumentalities" liability for underlying software capabilities "unduly burdens" AI innovation. * **The Dual Approach**: The FTC has largely ceased enforcing against AI capabilities, but has doubled down on classic Section 5 consumer protection enforcement against **"AI washing"** and unsubstantiated performance claims (e.g., **Click Profit**'s $20M+ judgment, **Workado**'s 53% vs. 98% accuracy settlement, and **accessiBe**'s $1M settlement). * **`illinois-workplace-ai-notice-regulations-hb3773`** (Illinois Postpones AI Workplace Notice Regulations, but Withdrawn Draft Reveals Broad Compliance Burdens) * **Finding**: On **June 2, 2026**, the Illinois Department of Human Rights (IDHR) temporarily postponed its rulemaking process and withdrew its draft rules implementing **HB 3773** (which took effect January 1, 2026). * **The Draft Rules**: Despite the postponement, the withdrawn draft reveals a highly aggressive regulatory posture: defining "use" as any AI system that "influences or facilitates" decisions, extending joint liability to third-party tech vendors, and requiring highly detailed multi-channel notices (disclosing developer, product name, data collected, and accommodation rights). ### 3. Registered Watches * **`Illinois IDHR Re-introduction of AI Employment Notice Rules`** (one_shot, live_process): To track when the IDHR re-introduces and finalizes its administrative rules for HB 3773. * **`Florida AG v. OpenAI Motion to Dismiss or Procedural Ruling`** (one_shot, live_process): To track OpenAI and Sam Altman's response and the court's first major ruling in the Florida AG product liability lawsuit.