Clearon AI

Author: Clearon AI

  • Federal Court AI Orders Are Splitting Into Clear Patterns

    Federal Court AI Orders Are Splitting Into Clear Patterns

    Current snapshot: June 4, 2026.

    Federal courts are not moving toward one uniform AI rule. They are moving toward a patchwork. Some judges prohibit AI use in filings. Some require disclosure or certification. Others simply remind lawyers that Rule 11, candor, confidentiality, and sanctions rules still apply.

    That makes the practical rule simple: check the forum before filing, check the judge before drafting, and verify every AI-assisted citation, quotation, factual assertion, and legal proposition before it goes to court.

    1. No-use or near no-use orders

    The strictest orders do not merely require disclosure. They prohibit AI use for filings or memoranda.

    Judge Christopher A. Boyko of the Northern District of Ohio has a standing order stating that no attorney or pro se party may use AI in preparing any filing submitted to the court. Judge Sharon Johnson Coleman of the Northern District of Illinois similarly states in her standing requirements that parties may not use AI to draft memoranda or as authority to support motions.

    These orders are still the minority approach, but they matter because they show that some courts view AI-assisted drafting itself as the risk, not just unverified AI output.

    2. Disclose if AI was used

    The more common approach is disclosure. These rules do not necessarily forbid AI. They require the filer to say when AI was used and sometimes to identify the tool or the AI-generated portions.

    The Northern District of Texas now requires a brief prepared using generative AI to disclose that fact on the first page under the heading “Use of Generative Artificial Intelligence.” If the required disclosure is absent, the filing operates as a certification that no part of the brief was prepared using generative AI.

    Judge Michael M. Baylson of the Eastern District of Pennsylvania requires a clear factual statement disclosing AI use in papers filed in cases assigned to him and a certification that all citations to law or the record have been verified. The Southern District of California Bankruptcy Court uses a disclosure and certification form for generative AI use in pleadings, motions, and papers.

    3. Certify or verify the work

    Some courts focus less on whether AI was used and more on whether a human verified the final filing.

    The District of Nebraska’s local rule requires a certificate stating either that no generative AI was used or that a human verified all generated text, including citations and legal authority. The District of Kansas reminds lawyers and pro se litigants that AI-assisted filings remain subject to existing duties of candor and accuracy and warns that the court may strike filings, impose sanctions, or require sworn AI-use statements.

    This bucket is likely to grow because it fits comfortably with the existing professional-responsibility framework: AI can assist, but it cannot be the final authority.

    4. Treat AI output as an unverified source

    Some orders frame the problem as source reliability. The District of Hawaii’s General Order 23-1 treats AI-generated material as an unverified source and requires a declaration when counsel or a pro se party submits material generated by an unverified source.

    That framing is useful. It avoids treating AI as uniquely mysterious and instead places it beside other unverified material: useful as a lead, not good enough as filed authority unless checked.

    But Hawaii’s approach also creates a line-drawing problem. The order defines unverified sources to include AI-generated briefs and memoranda, along with online briefs or memoranda drafted by paid writers that are not tailored to a specific case. The risk is that lawyers may now have to decide not only whether they used AI, but whether a particular research output, template, summary, or purchased work product counts as an “unverified source.”

    The last paragraph of the order matters because it carves out ordinary legal research. The court says the order does not affect basic research tools such as Westlaw, Lexis, or Bloomberg, and that no declaration is required when the sources can be found on those tools. That is a sensible safe harbor for conventional citation checking, but it also raises practical questions as legal research platforms add generative AI features. If a lawyer uses a research platform’s AI summary, answer, or drafting aid, is the source the underlying case law, the research database, or the AI-generated synthesis?

    For lawyers, the safest reading is narrow: a source is not verified merely because it appeared inside a trusted platform. The underlying authority still has to be located, read, and checked. Hawaii’s final paragraph reduces friction for traditional legal research, but it should not be read as a blanket blessing for every AI-assisted feature embedded inside a legal research product.

    5. Protect confidential and proprietary information

    Other orders focus on confidentiality. Judge Stephen Vaden of the Court of International Trade requires disclosure when a filing contains text drafted with generative AI and a certification that the AI use did not disclose confidential or business proprietary information to an unauthorized party.

    This is the piece many lawyers miss. AI orders are not just about fake cases. They are also about what happens when privileged, confidential, sealed, trade secret, business proprietary, health, financial, or export-controlled information is entered into a tool that may store, train on, or transmit user input.

    6. Proposed Federal Rule of Evidence 707

    The most important federal rulemaking item is not a filing-disclosure rule. It is proposed Federal Rule of Evidence 707, which would address machine-generated evidence. The pending rules materials place proposed new Evidence Rule 707 on the December 1, 2027 track.

    That issue is different from AI-assisted drafting. Filing orders ask whether lawyers verified what they submitted. Evidence rules ask whether machine-generated evidence is reliable enough to be admitted.

    What lawyers should do now

    • Check district-wide local rules and judge-specific standing orders before drafting or filing.
    • Do not assume a general Rule 11 review is enough if the judge requires a separate disclosure or certificate.
    • Keep enough internal record of AI use to answer a court question without waiving privilege or exposing mental impressions unnecessarily.
    • Do not enter confidential, privileged, sealed, business proprietary, trade secret, protected health, financial, or export-controlled information into public AI tools.
    • Verify every citation, quotation, factual assertion, record reference, and legal proposition outside the AI tool.

    Help us keep the tracker current

    Clearon AI is tracking federal and state court AI orders, standing orders, local rules, protective-order language, and sanctions decisions. If you have found a court order on AI that is not reflected here, please send it through the Contact page.

    The most helpful submissions include the court, judge, date, docket number or rule number, a link to the order or PDF, and a short note on what the order requires. We will verify submissions against primary court sources before adding them to the tracker.

    Primary sources checked

  • Trump’s New AI Executive Order Turns Frontier Models Into a Cybersecurity Priority

    Trump’s New AI Executive Order Turns Frontier Models Into a Cybersecurity Priority

    President Donald Trump signed a new artificial intelligence executive order on June 2, 2026, and the center of gravity is clear: cybersecurity, critical infrastructure, and the most capable frontier models.

    The order, titled "Promoting Advanced Artificial Intelligence Innovation and Security," does not create a broad AI licensing regime. It expressly says it should not be read to authorize mandatory preclearance, licensing, or permitting for the release of new AI models. But it still gives the federal government a more formal role near the front end of model release: identifying high-capability frontier models and arranging early, secure access before those models are shared more widely with trusted partners.

    This is not a general-purpose AI rulebook. It is a national-security and cybersecurity order. Advanced AI is treated as both a defensive asset and a possible accelerant for cyber risk.

    What the Order Does

    Four pieces do most of the work.

    First, it directs federal cybersecurity leaders to prioritize AI-enabled cyber defense across national security systems, Department of War systems, and civilian federal government systems. Within 30 days, CISA, in consultation with OMB and other White House cyber and national-security officials, is directed to issue binding operational directives and other guidance where appropriate.

    Second, it creates an AI cybersecurity clearinghouse. Treasury, the Department of War through NSA, DHS through CISA, and the National Cyber Director are directed to form a voluntary clearinghouse with AI companies and critical-infrastructure operators. The goal is to coordinate vulnerability scanning, validation, remediation, and patch distribution.

    Third, it directs federal officials to develop a classified benchmarking process for advanced cyber capabilities in AI models. That process will help determine when an AI model should be treated as a "covered frontier model" under the order.

    Fourth, it calls for a voluntary framework under which AI developers can work with the federal government to determine whether models under development meet the covered-frontier-model threshold. Developers may then give the government secure access to covered models, with confidentiality, cybersecurity, insider-risk, intellectual-property, and nondisclosure protections, for up to 30 days before release to other trusted partners.

    That is the legal story for AI companies. The order does not say, "submit your model for approval." It says the federal government wants a structured way to spot advanced cyber capability, review certain models before broader trusted-partner release, and coordinate deployment where national cybersecurity interests are implicated.

    Why It Matters

    The order keeps the administration's pro-innovation posture, but it also shows where federal oversight is likely to harden first. Not around generalized consumer AI rules, at least not here. Around cybersecurity, national security, critical infrastructure, and model capability thresholds.

    That should get the attention of several groups.

    AI developers will need to assess whether their model-development processes can support secure government engagement without compromising trade secrets, release timelines, or customer commitments. Even a voluntary framework can become practically significant when major labs, cloud platforms, federal contractors, or critical-infrastructure vendors are involved.

    Federal contractors and regulated entities should watch the CISA and OMB guidance that follows. The order directs action on federal systems, but it also points to access for state and local authorities and operators of critical infrastructure, including rural hospitals, community banks, and local utilities. That suggests downstream cybersecurity expectations may reach beyond Washington.

    Legal and compliance teams should also pay attention to the documentation burden. If a model could plausibly fall within a classified benchmarking process, companies will want a defensible internal record of model capabilities, cyber-risk testing, access controls, deployment plans, and third-party release decisions.

    The Frontier-Model Piece

    "Covered frontier model" may be the most consequential phrase in the order.

    The order directs federal officials to build a classified benchmarking process to assess advanced cyber capabilities and identify the threshold for that designation. The designation decision is assigned to NSA leadership in consultation with the National Cyber Director, the Assistant to the President for Science and Technology, CISA, and Department of War representatives.

    That approach keeps the most sensitive capability assessment out of public view. It also means companies may never get a clean public checklist for what makes a model covered. They may instead be dealing with a government-facing process built around classified benchmarks, agency judgment, and secure communications with federal officials.

    From a legal-risk perspective, this creates several practical questions:

    • How will a company determine whether to initiate voluntary engagement?
    • What internal evidence should support the company's view that a model is or is not likely to meet the threshold?
    • How will pre-release access be governed contractually?
    • How will intellectual property, model weights, system prompts, evaluations, logs, and vulnerability findings be protected?
    • What happens if a company disagrees with the government's assessment?

    The order does not answer those questions. It starts the process that will create them.

    Not a Licensing Regime, But Not Nothing

    The anti-licensing language is not throwaway. It appears designed to reassure industry that the administration is not recreating a mandatory pre-release approval system for frontier AI.

    But legal teams should not mistake that reassurance for irrelevance. Voluntary frameworks can still shape market expectations, procurement preferences, liability arguments, insurance underwriting, and board-level risk controls. If the federal government creates a recognized process for secure early access and frontier-model cyber benchmarking, companies that ignore it may eventually have to explain why.

    That is especially true in sectors where AI models are deployed into cybersecurity products, vulnerability detection, incident response, financial services, health systems, utilities, or other sensitive environments.

    The Altman-Musk Divide

    The industry's early reaction shows why that anti-licensing language was probably necessary.

    OpenAI has publicly embraced the final order's basic structure. Sam Altman reportedly said the order "gets the balance right," and OpenAI's chief global affairs officer, Chris Lehane, framed the issue as one for democratic institutions, technical experts, and public stakeholders. That fits OpenAI's broader posture: accept government-informed safety testing and standards for high-capability systems, while resisting a regime that turns every major model release into a permission slip.

    Elon Musk and xAI appear to be in a different, more skeptical lane. Axios reported that Musk, along with Meta's Mark Zuckerberg and White House AI adviser David Sacks, spoke with President Trump before an earlier version of the order was delayed. The final version that emerged was narrower: voluntary rather than mandatory, built around a 30-day pre-release access window, and explicit that it does not authorize preclearance, licensing, or permitting for new AI models.

    That does not mean xAI is rejecting federal testing. In May, xAI, Google, and Microsoft agreed to give the federal AI Safety Institute, now CAISI, access to models for security testing before release. The better reading is narrower: xAI appears willing to participate in government model testing, while the Axios reporting suggests Musk was part of the industry pushback against a heavier pre-release review regime.

    For legal teams, that distinction is useful. The frontier labs are not simply dividing into "regulated" and "unregulated" camps. They are drawing boundaries around the legal character of the process: voluntary cooperation, safety benchmarking, and secure government access on one side; mandatory licensing, public approval gates, and open-ended release delays on the other.

    Enforcement Against AI-Enabled Cybercrime

    The order also directs the Attorney General to prioritize enforcement against people who use AI to unlawfully access or damage computer systems, steal data, or facilitate other crimes. It specifically references federal computer crime and fraud statutes, including 18 U.S.C. 1028, 1030, and 1343.

    That section is short, but it does some work. It frames AI-enabled cyber misuse as an enforcement priority rather than a wholly new legal category. The administration appears to be saying that existing criminal laws already reach many AI-assisted cyber offenses, and DOJ should treat AI use as a reason to prioritize those cases.

    Companies should read that as a controls issue. AI agents, autonomous scanning tools, security research workflows, and employee use of AI in technical environments all need clear authorization boundaries. A tool that accelerates defensive work can also create evidence problems if it is used the wrong way.

    What To Watch Next

    The next 30 to 60 days will tell us more than the headline did.

    CISA guidance and any binding operational directives will show how federal agencies are expected to use AI-enabled cyber tools and whether contractors will see new expectations in security programs. Treasury, NSA, DHS, and the National Cyber Director's clearinghouse work will show how much private-sector coordination the government can realistically achieve. The classified benchmarking process will determine whether "covered frontier model" becomes a narrow national-security category or a broader marker for advanced AI cyber capability.

    The order is not a comprehensive AI law. It is not a privacy law, a copyright law, or a civil-liability framework. It does show where federal AI governance may harden first: cybersecurity.

    For AI companies and the organizations that rely on them, the practical takeaway is direct: model capability, cybersecurity readiness, release governance, and critical-infrastructure impact now belong in the same review process.

    Editorial Notes

    Suggested dek: The June 2 order does not create a mandatory AI licensing system, but it does create a federal path for frontier-model cyber benchmarking, secure early access, and AI-enabled cyber defense.

    Suggested social: The new AI executive order is not a broad licensing regime. It is something more targeted: a cybersecurity and national-security framework for frontier-model capability, pre-release access, and critical infrastructure defense.

    Related follow-ons:

    • What AI companies should document before engaging with the voluntary frontier-model framework.
    • Why CISA's next AI guidance may matter more than the executive order itself.
    • How AI-enabled cybercrime enforcement could affect companies using autonomous agents.

    Sources

  • Weekend Legal AI Roundup: What lawyers should catch up on Monday

    Weekend Legal AI Roundup: What lawyers should catch up on Monday

    The weekend did not produce a flood of legal AI news, but it did leave a few developments worth carrying into Monday.

    The biggest late-Friday carryovers were a new statewide Florida court rule on AI-assisted filings and a clear Big Law signal that Kirkland & Ellis wants to build more of its own AI infrastructure instead of renting all of it.

    There was also a narrower but still relevant enforcement development: the FTC has begun rolling out its TAKE IT DOWN Act enforcement channel. That is not the lead enterprise-AI story of the week, but it belongs on the synthetic-media and platform-obligations watchlist.

    Here are the items worth catching up on before the week gets moving.

    Florida put a statewide rule around AI-assisted court filings

    The Florida Supreme Court issued a May 28 administrative order and companion rule amendments replacing circuit-by-circuit AI disclosure requirements with a single statewide standard.

    The new framework puts the emphasis on something much more practical than generic AI panic: lawyers remain responsible for the existence and accuracy of cited legal authorities, and courts have express sanctions language to back that up. The change takes effect June 15, 2026.

    This is one of the cleaner signs yet that courts are moving from scattered warnings to operational AI-use rules. Florida is not banning AI. It is doing something more durable: turning verification, supervision, and filing discipline into a statewide workflow expectation.

    That matters to litigators, supervising partners, and in-house teams that review outside-counsel AI policies.

    The practical Monday-morning takeaway is simple: if your team uses AI in drafting or cite-checking, this is a good week to confirm who verifies authorities, how that verification gets documented, and whether your written AI-use guidance still sounds abstract when the court rule now sounds concrete.

    Kirkland is spending like AI infrastructure is now a strategic asset

    Reuters reported on May 28 that Kirkland & Ellis plans to spend $500 million over the next three to four years building a proprietary AI platform, with $100 million expected in 2026 alone.

    The report says the firm will still license some outside tools, but the headline point is hard to miss: one of the world’s largest firms appears to think the long game is not just buying AI products, but owning more of the workflow layer itself.

    This is a stronger market signal than yet another vendor demo or partnership announcement. If elite firms are willing to treat AI as internal infrastructure, that sharpens the build-versus-buy question for everyone else.

    It also reinforces a trend Clearon has been tracking for weeks: competitive advantage may sit less in raw model access and more in governed context, firm-specific knowledge, integration, and control.

    The practical Monday-morning takeaway is that law firms and legal departments evaluating AI tools should ask a more serious architecture question than whether a feature looks useful. The better question is which capabilities belong in a vendor stack, which should sit behind internal controls, and what gets harder to unwind once workflow, precedent, and usage data start concentrating in one place.

    The FTC’s TAKE IT DOWN rollout is not a core enterprise-AI story yet, but it is worth watching

    The FTC announced that it has begun enforcing the TAKE IT DOWN Act and launched a complaint channel for failures to honor valid removal requests involving nonconsensual intimate imagery, including AI-generated abuse scenarios described in the agency’s rollout.

    This is not the lead item for most law firms or in-house AI governance teams, but it is a real compliance signal for platform operators, trust-and-safety counsel, and anyone tracking synthetic-media obligations.

    It also shows how fast AI-specific legal questions can get folded into ordinary enforcement machinery once a law is in place.

    The practical Monday-morning takeaway is that if your organization operates a platform, moderation workflow, or user-generated-content channel, this is a useful prompt to review takedown intake, escalation paths, and whether synthetic-media response procedures are documented well enough to survive regulator scrutiny.

    What to watch this week

    Watch whether Florida’s court-rule move gets copied elsewhere, and whether more legal organizations start talking openly about AI as infrastructure rather than software.

    The recurring question is getting clearer: who controls the workflow, who verifies the work, and where responsibility actually sits once AI is inside legal operations.

  • Legal AI Roundup: This Week’s Pressure Points

    Legal AI Roundup: This Week’s Pressure Points

    This week's legal AI story was not the volume of headlines. It was the sharper pressure points underneath them.

    Copyright plaintiffs kept pushing AI disputes deeper into court. The FTC reminded the market that fake AI capability claims can become an enforcement problem quickly. And on the legal-workflow side, vendors kept leaning into a more practical message: the value is not only model output, but whether AI can operate inside governed institutional context.

    Here are the developments lawyers should know before the weekend.

    CNN sues Perplexity over alleged copying and distribution of news content

    CNN sued Perplexity in the Southern District of New York, alleging that the company unlawfully crawled, scraped, copied, and distributed more than 17,000 CNN stories, videos, images, and other works to power its products. The complaint also includes trademark allegations tied to supposed affiliation and premium-access claims.

    This pushes the publisher-AI conflict further into answer-engine behavior, output substitution, and source-rights questions tied to real-time content use. It also appears to be the first AI copyright case brought by a television network, which broadens the plaintiff set beyond newspapers, authors, and music-rights holders.

    The practical takeaway is that legal teams evaluating answer-engine or retrieval-heavy AI products should stop treating content provenance as a background issue. If the product experience depends on scraping, summarizing, re-serving, or commercially repackaging third-party content, rights questions are part of the product risk, procurement risk, and litigation-risk analysis from day one.

    Disney’s AI copyright case against MiniMax survived its first major dismissal push

    Judge Stanley Blumenfeld Jr. denied MiniMax’s motions to dismiss for lack of personal jurisdiction and failure to state a claim in the Disney-led copyright case over the Hailuo AI system. The ruling keeps the case alive and requires MiniMax to answer the complaint.

    Procedural rulings like this are easy to underrate, but they matter. They show that some courts are willing to keep AI copyright disputes moving rather than resolving them at the threshold. That means litigants, vendors, and enterprise users should expect more record development around training inputs, model behavior, distribution theories, and rights defenses before the law settles.

    The bigger signal is not that plaintiffs have already won. They have not. It is that courts may be prepared to let these cases mature long enough to produce more meaningful guidance. For legal and business teams, that means dataset governance and vendor diligence still belong on the live risk list, not in the category of speculative future problems.

    The FTC’s latest AI case is really about fake capability claims and bad consent stories

    The FTC announced settlements with Cox Media Group, MindSift, and 1010 Digital Works over claims that they marketed an AI-powered “Active Listening” advertising product that supposedly captured consumer conversations from smart devices and relied on consumer opt-in. According to the FTC, the product did not actually work that way, did not use voice data at all, and was instead built around resold email lists.

    This is a strong enforcement reminder that AI risk is not limited to model outputs or hallucinations. Marketing claims, technical representations, and consent narratives can create liability on their own. For in-house legal teams, that reaches product marketing, vendor diligence, privacy review, procurement, and internal signoff processes.

    This is exactly the kind of case that should make lawyers ask harder basic questions before any AI product or vendor pitch goes out the door: what does the system actually do, what evidence supports that claim, what data does it really use, and is the consent story real or just sales gloss? A lot of AI governance work still comes down to old-fashioned substantiation discipline.

    Harvey and DeepJudge pushed the legal AI market further toward institutional knowledge grounding

    Harvey and DeepJudge announced a partnership aimed at bringing prior work, negotiated positions, internal expertise, permissions, and ethical-wall-aware institutional knowledge directly into AI workflows. The pitch is not just better outputs in the abstract. It is AI that reflects how a specific firm or legal department actually works.

    This is one of the clearest workflow signals of the week. Legal AI value is increasingly being framed around governed context, access controls, precedent reuse, and organization-specific judgment rather than generic model performance alone. That matters for firms and law departments trying to separate impressive demos from systems they can actually supervise and trust.

    The strongest legal AI vendors are converging on the same message: the hard part is no longer just generating text. It is controlling what institutional knowledge the system can reach, how permissions carry through, whether outputs reflect the team’s own standards, and how that whole process stays auditable. That is a much more serious buyer conversation than a feature checklist.

    Why this week mattered

    The pattern this week was not just more legal AI news. It was more evidence that the important legal AI fights are getting more concrete.

    On one side, courts and plaintiffs are pushing harder on content rights, copying, and distribution theories. On another, regulators are reminding companies that exaggerated AI claims and sloppy consent narratives can still trigger ordinary enforcement tools. And inside legal workflow itself, vendors keep moving toward governed context and institutional knowledge as the place where durable advantage may actually sit.

    For lawyers, the through-line is straightforward: the important question is not whether AI remains exciting. It is whether the systems being bought, deployed, or defended can survive scrutiny around rights, representations, governance, and operational control.

    Related Clearon reading

  • Institutional Knowledge May Be Legal AI’s Main Competitive Layer

    Institutional Knowledge May Be Legal AI’s Main Competitive Layer

    The Harvey-DeepJudge partnership offers a clear picture of where legal AI is heading next: toward institutional knowledge.

    Harvey brings the workflow layer. DeepJudge brings prior work, negotiated positions, internal expertise, and permissions-aware access to what a firm or legal department already knows. Put together, the pitch is simple: AI should do more than produce a plausible answer. It should reflect how the organization actually practices.

    What is actually at stake

    A lot of legal AI value will be won or lost here.

    If a system cannot reflect prior positions, accepted language, internal judgment, and ethical-wall-aware access rules, the output may be fast but still generic. Useful, maybe. Institutional, no.

    The deeper buyer question is shifting from model quality alone to whether the model can operate inside the knowledge, permissions, and standards that make a legal team distinctive.

    What the partnership signals

    Harvey and DeepJudge are betting that the next wave of legal AI will be less about raw model performance and more about context control.

    That means legal teams should pay closer attention to:

    • how AI reaches internal knowledge
    • whether permissions and ethical walls stay intact
    • how prior work informs drafting and analysis
    • whether outputs reflect firm-specific or department-specific standards

    The bigger shift

    This fits the same broader pattern visible across iManage, Harvey, Anthropic, and other legal AI players. The market is moving away from model quality alone and toward workflow ownership, governed context, and knowledge grounding.

    That may sound less flashy than another reasoning benchmark. It is also much closer to where real legal advantage lives.

    Practical guide: Legal AI Workflows: A Governance Checklist for Legal Teams

  • Governed Context May Be Legal AI’s Main Infrastructure Layer

    Governed Context May Be Legal AI’s Main Infrastructure Layer

    iManage's latest platform shift puts a spotlight on a layer that much legal AI coverage still underrates: governed context.

    At ConnectLive 2026, iManage described a platform built around a context fabric, AI-specific controls, agent monitoring, and MCP-based access to institutional knowledge. Strip away the branding and the message is simpler: legal AI infrastructure is not only the model. It is also the system that controls what the model can safely reach.

    Where this gets real

    For law firms and in-house teams, a good demo is not enough. If AI cannot reach the right knowledge, respect permissions, preserve confidentiality boundaries, and leave a reviewable trail, the polish of the answer does not matter much.

    Governed context deserves more attention than the phrase usually gets.

    • knowledge access
    • permissions
    • monitoring
    • auditability
    • workflow control

    What buyers should watch

    iManage is trying to own that layer. That is a sensible strategy, but buyers should still test the claims carefully.

    The real diligence questions are whether the controls are granular, whether agent activity is actually visible, and whether firms can connect multiple AI tools without losing control of client and matter boundaries.

    The bigger shift

    The legal AI market is moving away from “AI as a feature” and toward “AI as a workflow and knowledge infrastructure problem.”

    That may sound less exciting than model hype. It is also where the durable power probably sits.

    Practical guide: Legal AI Workflows: A Governance Checklist for Legal Teams

  • OpenAI Is Moving Into Government Legal Workflows Through Eudia

    OpenAI Is Moving Into Government Legal Workflows Through Eudia

    OpenAI's partnership with Eudia offers a useful clue about where legal AI is heading next.

    This is a workflow story more than a chatbot story. Eudia says the partnership is aimed at government legal and acquisition teams, combining OpenAI's models with Eudia's operating layer for regulated work.

    What matters here is not simply which model sounds smartest. It is who gets inside the workflow and becomes hard to replace.

    Government is where this gets real

    Government legal and acquisition work is where AI stops feeling like a novelty and starts looking like infrastructure.

    Once AI touches contracting, legal review, and mission-critical decisions, buyers need to ask harder questions about:

    • control
    • auditability
    • permissions
    • human review
    • vendor concentration risk

    Those are not side issues. They are the real product.

    What buyers should take from it

    The public announcement is still high level, and it does not answer every diligence question. But it is a useful signal.

    Frontier-model companies are not staying behind the curtain. They are moving into legal and acquisition workflows through specialized partners that already understand the operating environment.

    For legal and procurement teams, that means the smarter evaluation lens is no longer just model quality. It is whether the workflow around the model is governable, reviewable, and defensible.

    The bigger shift

    This is one more sign that legal AI is moving beyond the demo layer.

    The winners may not be the companies with the flashiest model. They may be the ones that control the workflow around it.

    Practical guide: Legal AI Workflows: A Governance Checklist for Legal Teams

  • Disney v. Midjourney and the Broader Copyright Question for AI Users

    Disney v. Midjourney and the Broader Copyright Question for AI Users

    Disney v. Midjourney makes the AI copyright fight more concrete.

    The case is about training data, but it is also about outputs that allegedly look too much like famous protected characters and franchise imagery.

    What the case is actually about

    Disney, Universal, and affiliated rights holders sued Midjourney in federal court in Los Angeles on June 11, 2025.

    The case is:

    • Case: Disney Enterprises Inc. v. Midjourney Inc.
    • Court: C.D. Cal.
    • Docket: 2:25-cv-05275
    • Status: pending

    The studios' position is straightforward. They say Midjourney was built using copyrighted works and that the service can generate outputs that are too close to protected characters and expressive elements. The complaint reportedly includes example prompts and output images involving well-known properties, which is part of why the case landed so clearly in public discussion.

    Two examples from the complaint show why the output issue is getting so much attention:

    Cropped complaint comparison image showing an alleged Midjourney Homer Simpson output beside Disney reference images.
    Cropped complaint comparison image showing an alleged Midjourney Homer Simpson output beside Disney reference images. Source: Complaint, Disney Enterprises Inc. v. Midjourney Inc., No. 2:25-cv-05275 (C.D. Cal.), page 32.
    Cropped complaint comparison image showing an alleged Midjourney Minions output beside Universal reference images.
    Cropped complaint comparison image showing an alleged Midjourney Minions output beside Universal reference images. Source: Complaint, Disney Enterprises Inc. v. Midjourney Inc., No. 2:25-cv-05275 (C.D. Cal.), page 51.

    Midjourney’s likely response is also familiar. Training is not the same as republishing a work. Not every prompted image is substantially similar enough to infringe. And not every reference to a known character, franchise, or visual style cleanly collapses into liability for the platform.

    That is why this case matters. Both sides are arguing about where the legal line sits when a model produces commercially useful images that unmistakably evoke existing protected expression.

    Can businesses use Midjourney images commercially?

    Midjourney’s published guidance says customers generally own the images and videos they create and may use them commercially, subject to its terms and plan requirements. For businesses with more than $1 million in annual gross revenue, Midjourney says a Pro or Mega Plan is required for commercial use.

    That contractual permission is only one part of the analysis. It does not guarantee that a particular output is noninfringing, that the user owns every element in the output, or that the output qualifies for copyright protection. Midjourney’s terms provide the service and assets on an “as is” basis, disclaim a warranty of noninfringement, and place responsibility for using or redistributing assets on the customer.

    For business use, the practical controls should include:

    • confirming that the account and subscription plan permit the intended commercial use;
    • screening prompts and outputs for recognizable characters, logos, protected expression, and other third-party rights;
    • retaining records of prompts, source materials, edits, and human review;
    • requiring additional clearance before using AI-generated images in prominent campaigns, products, or customer deliverables; and
    • reviewing vendor terms regularly because platform rules and protections can change.

    Commercial-use permission from the platform answers whether Midjourney permits the use. It does not answer whether a rights holder may challenge it.

    Related Clearon AI analysis: OpenAI copyright MDL and data governance and AI-generated code and copyleft risk.

    The bigger issue

    For companies, the issue is not just whether Midjourney wins or loses.

    It is whether the business has decided what level of copyright and brand-adjacent risk it is actually willing to accept when employees use generative AI in public-facing work.

    Many legal teams are comfortable saying obvious character replication is out of bounds. The harder question is the gray zone. Is the company willing to rely on a fair use argument if a marketing image is styled to evoke Disney, South Park, or another highly recognizable visual world? Is it comfortable arguing that a prompt drew on a style, not a protected work? Is it willing to defend that position after publication, in a customer campaign, or in court?

    That is the governance issue this case sharpens. Companies need a view on where they are comfortable being aggressive, where they want to be conservative, and which arguments they are actually prepared to stand behind if challenged.

    They also need to account for contract risk, not just copyright doctrine. Most, if not all, major AI image providers put the user on the hook for at least some infringement risk tied to prompts, inputs, or outputs. Even when a vendor offers limited indemnity, it is often narrow and conditional. So a company deciding to operate in the gray zone may also be deciding that it, not the service provider, will carry much of the downstream claim risk.

    The Clearon AI takeaway

    Disney v. Midjourney turns AI copyright risk into a risk-allocation question for users, not just model developers.

    The practical lesson is less “never touch this” and more “decide, in advance, which copyright arguments your company is truly willing to own.”

    Sources

  • Weekend Legal AI Roundup: What Lawyers Should Catch Up On Monday

    Weekend Legal AI Roundup: What Lawyers Should Catch Up On Monday

    The legal AI signal over that weekend was not a new rule or a splashy lawsuit. It was a clearer market picture.

    The biggest vendors were moving closer to legal-specific workflow ownership, clients were getting louder about expecting real AI adoption, and the risk conversation kept shifting from abstract ethics to privilege, supervision, and workflow design.

    OpenAI is reportedly planning a legal-specific AI offering

    Here are the items worth knowing before the week gets moving.

    Artificial Lawyer reported on May 18 that OpenAI is planning a legal offering that could be branded as “Codex for Legal,” with legal-tech hiring and a vertical strategy similar to the company’s broader “Codex for almost everything” push.

    What it means for lawyers:

    If accurate, this is another sign that major model providers do not want to sit behind generic chat interfaces forever. They want to move into legal-specific workflow, tool integrations, and day-to-day lawyer environments. That raises practical buyer questions about lock-in, governance, and how much of the legal work surface gets controlled by a handful of platform vendors.

    Practical takeaway Monday morning:

    Legal teams should treat this as a market-structure development, not just another product rumor. If your organization is evaluating legal AI tools, ask where workflow control is heading, what data leaves your environment, and how easily you could switch tools later.

    Source: Artificial Lawyer, May 18, 2026.

    Anthropic’s legal play is getting harder to dismiss as a side experiment

    What happened:

    A May 16 Artificial Lawyer analysis of Anthropic’s latest Claude for Legal webinar described a platform that now has 12 legal plugins, customization options, MCP connectors, and a clear push to stay inside the lawyer’s working environment, especially around Microsoft Word and related tools.

    What it means for lawyers:

    The more interesting point is not plugin count. It is the strategic direction. Anthropic appears to be competing to become part of the lawyer’s primary workspace rather than a bolt-on drafting helper. That has consequences for procurement, document governance, supervision, and training because the tool starts to shape how legal work is actually done.

    Practical takeaway Monday morning:

    If you are buying or piloting legal AI this quarter, compare products on workflow fit and governance controls, not just answer quality. The winning tool may be the one that best controls document flow, user permissions, and auditability.

    Source: Artificial Lawyer, May 16, 2026.

    One Friday item that still matters Monday: clients are openly warning firms not to lag on AI

    What happened:

    In a May 15 Law.com Corporate Counsel Q&A, Salesforce chief legal officer Sabastian Niles said firms that fail to embrace AI risk losing efficiency, talent, and clients.

    What it means for lawyers:

    This is the client-pressure version of the legal AI story. The issue is no longer just whether firms can use AI safely. It is whether sophisticated buyers will start treating competent AI adoption as part of baseline service quality. That puts pressure on outside counsel to show not only that they use AI, but that they use it in a controlled, defensible way.

    Practical takeaway Monday morning:

    Law firms should be ready for more AI diligence from clients, especially around approved tools, data handling, supervision, and billing expectations. In-house teams should expect more firms to market AI capability and should separate real workflow maturity from demo-stage claims.

    Source: Law.com Corporate Counsel, May 15, 2026.

    The risk conversation is moving toward privilege architecture, not generic AI fear

    What happened:

    An ACC program scheduled for May 18 frames 2026 legal AI risk around privilege, discovery, professional responsibility, meeting notetakers, vendor diligence, and what it calls a defensible “privilege architecture.”

    What it means for lawyers:

    That framing is useful because it is more mature than blanket “don’t use AI” advice. The issue is no longer whether AI creates risk. Of course it does. The harder and more practical question is what legal workflow, vendor terms, access controls, and supervision rules let teams use AI without casually blowing confidentiality or evidentiary discipline.

    Practical takeaway Monday morning:

    This week is a good time to review whether your organization has an actual AI workflow policy for legal work, not just a general AI statement. Sensitive legal use should happen only in an approved enterprise environment with clear rules on prompts, retention, exports, and human review.

    Source: Association of Corporate Counsel program page, May 18, 2026.

    What to watch this week

    Watch whether the legal AI story keeps consolidating around workflow ownership. OpenAI, Anthropic, contract platforms, and enterprise clients all seem to be pushing in the same direction: less interest in standalone AI novelty, more interest in who controls the place where legal work gets drafted, reviewed, approved, and handed off.

    If that trend holds, the most important legal AI questions this week will not be which model is smartest. They will be who owns the workflow, what are the guardrails, and what happens to client trust when those answers are fuzzy.

  • The UK Is Moving Automated Decision-Making Away From the EU Model

    The UK Is Moving Automated Decision-Making Away From the EU Model

    Estimated reading time: 3 minutesTopics: Policy & Compliance · AI Governance · International AI Regulation · Legal Risk · In-House AI Adoption

    The UK's recent data-law changes matter for AI governance because they suggest a real break from the EU approach to automated decision-making.

    If you want the official legislation, the UK law is here: Data (Use and Access) Act 2025.

    Under section 80 of the Data (Use and Access) Act, the UK has replaced the old Article 22 framework with a more permissive structure: automated decision-making with safeguards, rather than a prohibition-first starting point.

    This is a real shift

    Under the classic Article 22 model, the analysis usually began with a restriction. The UK's newer approach is more operational and less categorical. The question becomes less "is this forbidden unless an exception applies?" and more "what safeguards, transparency, and review rights are required when this happens?"

    That may sound subtle, but it matters. It gives companies more room to deploy automated systems, while also increasing pressure to justify how those systems are used.

    What multinational teams should watch

    A lot of organizations still hope they can run one clean global policy for AI-enabled decision-making. The UK’s move makes that harder. If the EU and UK keep drifting apart here, legal teams may need separate assessments for profiling, scoring, and model-driven recommendations that affect individuals.

    That does not just affect flashy AI products. It can reach ordinary systems used in employment, insurance, financial services, fraud detection, customer eligibility, and prioritization workflows.

    The takeaway

    The UK is not abandoning regulation. It is choosing a different posture. A permission-with-safeguards model still requires governance, and in some ways it requires better governance because companies have more room to act.

    Cross-border AI compliance is starting to look less like one policy problem and more like jurisdiction management. That is the part legal teams should plan around now.