Most people assume that AI is coming to take jobs away from lawyers, doctors, and financial planners. That might happen in some cases, but the bigger story is being overlooked. The real winners will be the professionals who know how to delegate the boring parts of their job to AI, and use the extra time to focus on what they do best.
The truth is, despite their protestations and disclaimers to the contrary, tools like ChatGPT already offer some types of legal, medical, and financial advice that is better and cheaper than what many professionals provide. The problem is that AI platforms still pepper the good advice with just a dash of really terrible advice. The recent swath of AI disclaimers, warning not rely on AI to diagnose your appendicitis or move your retirement savings into Kardashian NFTs, indicates some naïve souls are still unaware of AI’s innate ability to make life-altering mistakes. Compounding the problem is that AI does not make the same types of mistakes humans make. This makes it nearly impossible for a non-subject-matter expert to separate the wheat from the chaff.
Still, it is clear where things are heading. The AI tools that will win are not the ones trying to replace your doctor or lawyer. The winners will be the tools that handle the behind-the-scenes tasks faster and more accurately than any human ever could.
If you are a professional, this is your cue.
Every job has two parts: the thinking and the doing. AI is getting better at the doing. Fast. That means the professionals who thrive will be the ones who use AI to automate the “doing” so they can spend more time thinking. More time talking with clients. More time spotting what AI will miss. More time being human.
That is what clients really want. They do not want to pay you $400/hr to fill out forms or run reports. They come to you for advice and vision they can trust. They come to you for judgment, empathy, and reassurance. Those are not things AI is replacing any time soon. But you need to give yourself the space to deliver them.
If you are still doing everything, the writing, the research, the scheduling, the billing, for yourself, you are not only wasting time, you are falling behind. The sooner you let AI handle the repetitive stuff, the more time you have to actually help people.
That is not just smart; it is the future of your profession.
ChatGPT just made a big change to its Usage Policy. As of October 29, 2025, OpenAI now prohibits users from using ChatGPT to give legal or medical advice. If you are relying on ChatGPT for help with court cases or medical conditions you are probably not reading this post, but if you are, you should probably stop. Not so much because it violates ChatGPT’s new Terms of Use, but moreso because it may easily end in disaster.
What Changed?
OpenAI updated its usage policy to make clear that users are not allowed to use ChatGPT to provide “tailored advice that requires a license, such as legal or medical advice, without appropriate involvement by a licensed professional.” That means if you are using ChatGPT to answer questions like “Should I sue?” or “What medication should I take?” you’re now violating OpenAI’s new policy. In addition to it possibly killing you, doing so could also result in suspension or banning of your account.
Why the Change?
Two main reasons: risk and responsibility. OpenAI likely made this change because of ongoing concerns that users were relying too heavily on ChatGPT for high-stakes decisions. Both legal and medical decisions can have serious consequences. And when the advice comes from a chatbot that sometimes makes things up, those consequences can be disastrous.
Lawyers and doctors go through years of training. ChatGPT doesn’t. The worst part is that even when it’s dead wrong, ChatGPT can be very convincing. Courts have already sanctioned lawyers for submitting AI-generated briefs with fake case law. And people have reported taking incorrect medical steps based on chatbot responses. OpenAI is simply drawing a clear line to reduce those risks, both for users and for itself.
What This Means for You
If you are a consumer, do not rely on ChatGPT for specific legal or medical advice. It is really that simple. General education is fine. But if your life or livelihood depend on it, talk to a real person, preferably one professionally licensed to give that advice. Your health and your legal rights are simply too important to irreparably damage by relying on well-crafted, but ultimately erroneous, advice.
If you are a lawyer or doctor, you should also not be using ChatGPT to give advice to clients or patients. Even if the chatbot gets some things right, that is not good enough. As the American Bar Association recently made clear, attorneys who use AI must personally verify its accuracy and protect ALL client data, not just some of it. The same goes for doctors under HIPAA.
What You Can Still Use ChatGPT For
OpenAI is not banning all legal or medical topics. You can still ask for help understanding broad legal concepts or general medical information. You just cannot rely on it for tailored advice. Think of it more like a textbook than a professional. You can learn about “what a patent is” or “how high blood pressure works,” but do not ask it to tell you what type of patent to file or what medication to take.
Final Thoughts
ChatGPT is a powerful tool. But like any tool, it has limits. The new ChatGPT policy serves as a reminder: some jobs still require human judgment. When your health or your legal rights are on the line, use the right tools, including lawyers and doctors who know what they are doing.
And if you are using ChatGPT in your business, now is the time to review your AI Acceptable Use Policy. Make sure everyone follows your internal AI policies as well as ChatGPT’s new usage policy (so you do not get banned from the platform), and make sure all of your employees know exactly what they can and cannot use AI for in your business.
Artificial Intelligence makes it easier than ever to fake someone’s voice, face, or entire identity. These AI-generated “deepfakes” can copy a person’s speech, movement, and image so convincingly that many people can’t tell the difference. That raises an important question: if someone makes a deepfake of you, how can you make them stop?
The Patchwork of State Laws
The problem is that deepfake protections are not consistent across the country. Some states have strong deepfake laws, some of which may even hamper legitimate businesses and online commerce. Others have none at all. There is no single federal law that clearly protects against unauthorized deepfakes. That makes enforcing your rights harder, especially when the person using a deepfake of you in another state. Depending on how the deepfake is used, you may have a cause of action under one of more of the following state or common law causes of action:
Right of privacy – Protects your private information
Right of publicity – Prevents commercial use of your image without authorization
Defamation – Prohibits false statements that damage your reputation
False light – Addresses emotional distress caused by a misleading impression of you
47 states have enacted some type of deepfake law. These laws typically cover one of five types of deepfake use:
Nonconsensual sexual deepfakes – May cover both adult and child images or videos
Political deepfakes – Prevents deceptive imagery, often right before an election
Digital replicas – California prohibits certain “digital replicas” of people without proper protections in place
Criminal laws – Pennsylvania criminalizes disseminating deepfakes with fraudulent intent
What About Federal Law?
Although there should be, there is currently no federal law preventing deepfakes. That means you may have to look to other laws to see if they fit your particular deepfake issue. Federal trademark law (like the Lanham Act) can sometimes protect people from false endorsements. But it mainly helps celebrities or people with commercially recognized identities. If you are not famous, federal law likely will not protect you from deepfakes. The FTC has also promulgated rules against deceptive endorsement, but it may be difficult to get the FTC to investigate a particular deepfake. You may also have a cause of action under state and federal unfair competition laws.
Take it Down Act – Forces websites to take down non-consensual imagery
Copyright law- If the deepfake is based on third-party copyrighted material
Why Deepfakes Are Different
Deepfakes created by AI can spread quickly and look real. Once they are online, they are hard to stop. AI also makes it cheap and easy to create fake content using your likeness, making the problem worse. Existing privacy, publicity, and defamation laws were not built to handle this kind of high-tech impersonation.
What Can Be Done?
In 2024, the U.S. Patent and Trademark Office (USPTO) held a roundtable on deepfakes. Speakers from the entertainment, sports, and tech industries called for stronger laws. They urged Congress to create clear national rules to protect people from deepfake misuse. The USPTO is now preparing recommendations for how lawmakers and regulators can better protect your identity in the AI age.
The Streisand Effect
As with certain types of defamation, filing a lawsuit may direct a lot of unwanted media attention at what was previously an isolated issue. Most deepfakes, while making an initial splash, slowly fade away in a short period of time, as the internet seeks the next new thing. The last thing you want to do is to fan a dying fire and keep your deepfake on everyone’s mind. While ignoring a deepfake may be tough to stomach, it may be a better idea than driving the entire internet to take a look at the deepfake that is causing you so much angst. Before you take any action regarding a deepfake, ask yourself what is your end goal, and whether ignoring the deep fake might be the best means to accomplish that goal.
What Should You Do Now?
If you are concerned about someone using your name, face, or voice in a deepfake:
Keep detailed time-stamped records of any unauthorized use
Contact an attorney with experience in deepfakes, defamation, and AI issues to determine which state and federal laws may protect you
Until new federal laws are passed specifically addressing deepfakes, the best protection is staying alert and getting legal help quickly.
Final Thoughts
Deepfakes are here, and they are only getting more realistic and more common. While existing state and federal laws offer some protection, they often fall short when it comes to AI-generated impersonation. Stronger laws are needed. But for now, understanding your existing rights under the various state and federal laws and acting fast may be your best defense.
If you are involved in content creation, AI development, or simply follow the big picture of the internet economy, the recent lawsuit filed by Reddit is something you should probably understand and follow. This lawsuit is about far more than one company suing another, it touches on who controls data, how AI gets trained, and what the rules for AI training will look like going forward.
What happened: Reddit sues over alleged large?scale scraping
On October?22,?2025, Reddit filed a lawsuit in New?York federal court against Perplexity AI and three web-scraping services: SerpApi (Texas), Oxylabs UAB (Lithuania), AWMProxy (former Russian botnet) and SerpApi (Texas).
Reddit accuses the web scraping defendants of:
Scraping billions of Reddit posts/comments without permission.
Circumventing anti?scraping protections by Reddit and Google by using Google search result pages as a back?door to Reddit content.
Violating laws including copyright, unfair competition, and unjust enrichment.
Reddit accuses Perplexity of ignoring Reddit’s cease?and?desist letter, sent May?2024, and then increasing references/citations of Reddit content by “forty?fold.”
In short, Reddit claims that the defendants treated its user?generated content as free training fuel for AI systems without a licensing agreement, discouraging other companies from entering into paid licenses with Reddit to use its content.
Why this matters: implications for data rights, AI training, and content platforms
Data control and monetization
Reddit’s case highlights that platforms may increasingly assert that user?generated content is not “free for anyone to use” at scale for commercial AI systems. Reddit argues that it already licenses its content to major players like Google LLC and OpenAI under agreements that include terms protecting Reddit.
If Reddit succeeds, content platforms will gain leverage in negotiations with AI companies in demanding higher compensation and/or tighter control over AI access to content.
Training data and “public data” assumptions
Many AI developers have assumed that publicly available internet content is fair game for training large language models or answer engines. Reddit’s suit challenges that assumption, especially when circumventing security measures to access user content currently licensed to third-parties.
Scraper liability and indirect access
Reddit alleges the defendants did not scrape Reddit directly, but scraped Google search result pages to harvest Reddit content (i.e., an indirect path). If the court awards Reddit damages for this type of indirect access through search engine results pages (SERPs), this could broaden liability for many other types of indirect access to content.
Precedent for AI business models
This lawsuit could set a new legal precedent, changing how AI companies build business models, especially those that rely on massive ingestion of web content without any kind of license. The outcome may force large AI companies to license more content, while squeezing smaller AI companies out of the market altogether.
Key Legal Claims & Technical Issues
Let’s break down the major legal and technical points at play.
Legal claims
Bypassing Reddit and Google Security: Reddit is claiming that the defendants have violated the Digital Millennium Copyright Act (DMCA), which prohibits anyone from circumventing technological measures controlling access to copyrighted works.
Unfair competition: Reddit claims the defendants’ access to Reddit content has gained for themselves an undue competitive advantage.
Unjust enrichment: Reddit argues that the defendants have been unjustly enriched through their access to Reddit content.
Civil Conspiracy: Reddit argues SerpApi and Perplexity have entered into agreements for the purpose of illegally circumventing Reddit’s technological control measures to gain access to Reddit content.
Technical / access issues
Violation of Google and Reddit terms of service.
Use of: proxies, fake user-agent strings, shifting IP addresses, bots to mimic human users, rate limit circumvention, and CAPTCHA circumvention.
Indirect content scraping via Google SERPS access Reddit content, rather than directly from Reddit’s API or website.
Increased citation volume after cease?and?desist, which Reddit takes as evidence of unauthorized use.
What this means for Reddit and the AI industry
For Reddit
Reddit is attempting to set a precedent for its licensing-of-user-generated-content-for-AI-training model.
Reddit may also set a precedent that SERPS scraping of user-generated content can be blocked and compensated.
Reddit is taking a shot across the bow of AI companies that if you want to train on Reddit’s third-party content, you have to pay to play.
For AI companies and content platforms
AI firms may have to take greater care how they obtain and use third?party content for training. “Publicly accessible” may not necessarily mean “free for AI training.”
Other platforms with valuable user?generated content may now demand licensing, or institute stricter API or access rules.
Scraper?services and data?broker services may find increased risk of liability if user-generated content platforms pursue them as enablers of unauthorized data acquisition.
A win for Reddit may spur more litigation and possibly state and federal regulation around training?data sourcing, fairness, transparency, and consent.
Enormous liability and/or license-fees may push smaller AI platforms out of the market and prevent new AI platforms from obtaining the capital needed to launch.
For content creators and users
Users who generate content on Reddit will still be left empty-handed when Reddit licenses their content to third-parties.
The “value” of online communities may become more visible and monetized rather than assumed to be free, possibly leading to new platforms developing a revenue-sharing model (similar to YouTube) for users who generate the content.
Platforms may implement more visible policies about how user content may be used in AI training.
Outlook: Watch Points and What to Monitor
Here are the key developments to watch:
Court rulings: How the court handles these legal claims: decisions on scraping via search results, indirect access, circumventing technological security measures, unfair competition, and the role of robots.txt/crawling protections.
Licensing agreements: Whether more platforms follow Reddit’s path and demand paid licenses for AI training data.
AI company responses: Will AI developers change their data?acquisition strategies, increase transparency, or negotiate deals proactively?
Regulatory action: Legislators or regulators may step in to clarify rights around data, training, consent, and attribution for AI systems. Federal legislation is needed to preempt a hodgepodge of ill-considered state laws.
Technical safeguards: Platforms may strengthen anti?scraping measures, auditing of access logs, and contracts with data brokers and scrapers.
Closing Thoughts
The Reddit v. Perplexity lawsuit is not just another tech dispute. It could reshape how platforms resell your content, how AI systems are trained, and how the internet handles user?generated content in general. If you are a creator, business owner, platform manager, or AI developer, you would be well-advised to follow this case closely.
If you have already filed a patent application, you likely want to get it examined as soon as possible. Fortunately, the United States Patent and Trademark Office (USPTO) has introduced a new option to help you do exactly that. In this post I will walk you through how to take advantage of this option, and other best practices to help accelerate your application’s examination.
What’s new: The Streamlined Claim Set Pilot Program
In October?2025 the USPTO published a notice for its Streamlined Claim Set Pilot Program. According to this official notice:
The pilot is designed for patent application that have already been filed.
Qualifying applications will be advanced out of turn (i.e., accorded special status) for examination until a first Office action is issued.
To qualify, applications must have, or be amended to have, no more than one independent claim, no more than ten total claims, and no multiple dependent claims.
To qualify you must electronically file a petition through the Patent Center using form PTO/SB/472.
The program will run for up to 12?months from its publication date or until each Technology Center has docketed about 200 eligible applications, whichever comes first.
In short: If you can keep your claim set tight and file the required petition, you may get your application bumped ahead of the queue.
Why this matters
Normally, patent applications are examined in the order of their filing date (or national stage entry date).
This pilot gives applicants a way to cut in line, at least up to the first Office action, which can mean faster feedback and reduced pendency.
Getting earlier feedback means you can respond sooner, refine your strategy earlier, and potentially start generating revenue from your invention more quickly.
From a capital raise perspective, getting faster examination could persuade investors to invest more money faster.
For portfolio planning and licensing discussions, moving your application ahead can improve the timing of your patent?asset development.
Who is eligible
You will want to check carefully if your application qualifies. Here are the key eligibility points:
Must be an original, non?reissue, non?continuing utility application filed under 35?U.S.C. §?111(a).
The application must have an actual filing date before the publication date of the notice (October?27,?2025) in the Federal Register.
National stage applications under 35?U.S.C. §?371 are not eligible.
The claim set must satisfy: no more than one independent claim; no more than ten total claims; no multiple dependent claims; all dependent claims must properly refer to previous claims and recite all limitations.
If your application included a non?publication request, you must rescind it at or before the time you file the petition to make special.
The specification, claims, and abstract must have been submitted in DOCX format (as required for electronic filing) at the time of filing.
The inventor (or joint inventors) must not have been named on more than three other nonprovisional applications in which a petition to make special under this pilot program has been filed.
Practical Steps to Apply for the Pilot Program
Here’s a step?by?step checklist to help you move your application toward the front of the line:
Evaluate claim set size
Review your current claim set. If you have more than one independent claim, more than ten total claims, or any multiple dependent claims, then you must amend the application (for example via a preliminary amendment) so that the claim set matches the pilot program’s requirements.
If you file an amendment to reduce claim count, make sure you do so before or with the petition. Otherwise, you may not meet the eligibility restrictions.
Prepare petition form PTO/SB/472
Fill out the certification and petition to make special under the pilot program.
Ensure that your filing is electronic, through the USPTO’s Patent Center.
Pay the required petition fee under 37?CFR?1.17(h).
Rescind non?publication request if needed
If your application previously requested nonpublication, file a rescission (using form PTO/SB/36) no later than the time you file the petition.
File the petition timely
You must file the petition before issuance of the first Office action or written restriction requirement. If the application has already been docketed to an examiner, the petition will generally be dismissed.
Make sure the application was filed via the USPTO’s electronic filing system and the specification, claims, abstract were DOCX compliant at filing.
Monitor docket status
Once accepted, the application will be placed on a “special docket” until the first Office action is issued. After that, the application falls back to the regular examination queue.
Keep track of whether each Technology Center is still accepting applications for the program (since acceptance may close early after 200 applications have been fast-tracked under the pilot program).
Respond carefully to the Office action
Because the application was accelerated only up to the first action, your response to that first action matters. A fully responsive reply helps maintain progress.
If your amendment adds claims that violate the pilot program requirements (e.g., more than one independent claim), the examiner may treat the amendment as non?compliant and revert you to the regular queue.
Other Ways to Accelerate Examination
Even if you cannot qualify for the pilot, here are other tools to help move your application ahead:
Prioritized Examination (Track One): This is another USPTO program offering faster examination (typically within about 12?months) for a fee.
Petition to Make Special under other rules: For example, you may qualify for a petition to make special if you are over 65 years-old or if your invention relates to the quality of the environment, development of energy resources, or fighting terrorism.
Strategy in claim drafting: Keeping claims clear, focused, and limited in number may not automatically guarantee “front of the line” status, but it tends to reduce delays caused by restriction requirements or claim?set rejections.
Dialogue with the examiner / early interview: Engaging with the examiner, addressing issues directly, and holding high-quality interviews with the examiner can speed up the process.
Amendments to simplify your application: While you don’t want to sacrifice scope, trimming extraneous claims and narrowing dependencies may reduce complexity and thus examination time.
Tips and Best Practices
Plan early: If you think you may qualify for this pilot program you should act quickly, before the Technology Center assigned to your application accepts 200 pilot applications. If you are not eligible for this program, but still wish to take advantage of one of the other fast-track options, plan your claim strategy and drafting with your desired fast-track option in mind.
Avoid multiple dependent claims if you want to qualify for this pilot program, multiple dependent claims are a non-no.
Keep claim count low: Even outside this pilot, fewer claims often mean less examiner time and fewer complexity issues.
Maintain DOCX compliance: Make sure your specification, abstract, and claims are in DOCX format when filed. That is a requirement for the pilot.
File the petition timely: Filing the petition after an Office action is issued or after your application is docketed to an examiner will disqualify you for this pilot.
Budget wisely: The petition fee and possibly extra costs for claim amendments should be included in your prosecution budget (although reducing the number of claims could easily offset the overall prosecution cost by more than the fees required to get the application fast-tracked).
Monitor changes: Since this is a pilot program, the USPTO may modify or terminate it early (for example, once a threshold number of applications is accepted in a Technology Center).
Educate inventors and stakeholders: While the pilot program can speed up the initial examination and commercialization, after the first Office Action, your application goes back into the regular examination rotation.
Work with your patent counsel: These eligibility rules are nuanced. A patent attorney can help you evaluate whether your application qualifies and prepare the petition correctly.
Final Thoughts
If you are looking to accelerate your patent application, the Streamlined Claim Set Pilot Program provides a clear path to move your application ahead (subject to meeting specific requirements). By keeping your claim set tight, filing the petition properly, and responding quickly, you may arrive at the first Office action faster.
Even if you do not qualify for this pilot, the best practices of streamlined claiming, early examiner engagement, and strategic prosecution remain effective. Incorporate these into your patent?strategy playbook and you’ll move closer to the front of the line.
Artificial intelligence (AI) is no longer a theoretical business tool. It is a real tool your employees are likely using today. Whether it is for writing emails, summarizing documents, or analyzing data, AI is quickly becoming a daily part of business. But without clear rules in place, it is also becoming a serious legal and security risk.
That is where an AI policy comes in.
What Is an AI Policy?
An AI policy is a set of rules telling your employees how and when they can use AI at work. It should cover what tools they can use, what kind of data they can input, what they can do with the output, and what to do in case a problem arises.
Without an AI policy, employees might put private customer data into public AI tools. They might use AI to make business decisions without understanding the risks. Or worse, they might publish AI-generated content that violates someone’s copyright or privacy rights.
Why Do You Need One?
1. To protect your data. Most public AI tools learn from the data you give them. If one of your employees pastes a confidential contract into an AI tool, that contract could end up in someone else’s AI-generated document tomorrow.
2. To avoid breaking the law. AI tools are powerful, but they’re also unpredictable. They can generate false or misleading information. If you rely on that information to make decisions or share it with others, you could be held responsible.
3. To control who owns the work. Unless your employee changes the AI output in very specific ways, your company may not legally own it. That could mean losing control over marketing materials, code, or other creative work you thought belonged to you.
4. To manage employee use. Without trackable login credentials and documented use, you cannot monitor who is using AI or how they are using it. That opens the door to misuse, inefficiencies, and even legal exposure.
5. To limit liability. Improper use of AI can cause catastrophic damage for which your company may be liable. Putting all employees on the same page, when it comes to appropriate use of AI, reduces the chance that you find your company on the angry end of a large lawsuit.
What Should an AI Policy Include?
Depending on how big your company is and what it does, your AI policy may include one or more of the following clauses:
Approved AI platforms and use cases
Privacy and data protection rules
Periodic system assessments for accuracy, reliability, and security holes
Risk management protocols
Human review and editing of AI content before dissemination outside company
Employee training requirements
Compliance mandates to comport with changing laws governing AI use
Respect for third-party copyrights
Ownership rules for AI-generated materials
Required employee training and consent
Address industry specific laws – HIPAA, CCPA, GLBA, COPPA, etc.
Designate Data Protection Officer and duties
Unique AI usage credentials
Issue reporting procedures
Human oversight and consequences for breaking the rules
A list of approved AI tools for approved uses
Time periods to review and revise the AI policy
Customize Your AI Policy
When it comes to AI policies, there is no one-size-fits-all solution. Different companies use AI in different ways, which means they all need different AI policies to address these differences. Whereas smaller companies may not require a dedicated Data Protection Officer, very large companies may require an entire data protection department. Most importantly, do not include anything in your data protection policy that you do not intend to enforce. Including mandates in your data protection policy that you refuse to enforce may actually backfire on you, exposing you to even more liability than if you had simply not addressed that particular issue at all.
The Bottom Line
Implemented correctly, AI can save your company time and money, but it can also create serious problems if used the wrong way. An AI policy helps you extract the benefits of AI while avoiding the risks. Most importantly, an AI policy, customized for your company, places everyone in your company on the same page when it comes to proper use of AI. This reduces the likelihood of misuse and the potential catastrophic legal liability that comes with it.
If you do not have an AI policy, now is the time to start. The sooner you start setting out clear rules for AI use, the better protected your company will be.
Artificial Intelligence (AI) is changing the way we work and live. But instead of helping businesses make the most of this new tool, states like California, Texas, Utah, and Colorado are putting up roadblocks. Their new AI laws are confusing, expensive to follow, and could stop companies from using helpful AI tools to increase productivity.
Businesses Are Scared to Use AI, for Good Reason
I recently spoke with manufacturers at a large business conference after a presentation on these new AI-focused state laws. Each national manufacturer I spoke with was concerned about how these new laws might affect the implementation of AI into their businesses. They were worried that using AI might get them into legal trouble, even though no one knows exactly what these new laws mean or how courts might interpret them. Not surprisingly, none of these business owners wanted to be the first to test these new laws in court.
Why These Laws Don’t Work
The goal of any law should be to solve problems without causing new ones. But these new state laws do the opposite. These states are creating a patchwork of rules that do not even match up with each other. Each of these states has its own language and its own set of requirements.
Imagine trying to run a business that uses AI across 50 states. You would have to hire lawyers to:
Read and explain every single law
Update your systems to meet different rules in each state on an ongoing basis
Take an educated guess as to how courts may interpret these new laws
Propose policies to avoid breaking any of these ambiguous, untested laws
Defend you if one of those states decides your AI tool breaks its law
That is a lot of money and a lot of time. And none of it helps businesses grow or make better products. Instead, it keeps lawyers busy and businesses nervous.
California’s New Law: A Case Study in Overreach
Let’s look at just one of these new laws, California’s SB243. It sounds good on the surface—trying to protect kids from harmful AI content. But in practice, it is far too broad. The law applies even to general tools like ChatGPT, requiring operators to use “evidence-based methods” to detect suicidal thoughts. That may sound good in theory, but what are those “methods”? And how do you apply them in a way that does not limit up the experience for the 99% of users who are not at risk?
The new law also requires AI companies to:
Notify crisis service providers
Detect and remove harmful content
Report every year on their efforts
Follow vague rules on what to do if the user might be a minor
Who decides which online user “might” be a minor? How do you make that call without violating someone’s privacy? How much will this cost? How much of a burden does this place on current users? And how is a startup supposed to afford this on top trying to comply with all of the similar, but different, laws from other states?
The Real Cost of Compliance
The more states that pass these kinds of laws, the more money companies have to spend just to comply with all of these new laws. That is money they could be using to make better products, hire more people, and/or cut prices. Instead, these companies are hiring lawyers and compliance officers to deal with confusing rules that could change at any moment.
Worse, these laws are helping big tech at the expense of startups. Big companies have the money to keep up with state-by-state regulations. Small companies do not. That is unfair, and not how you grow innovation.
These Laws Are Not Like Property Laws
Some argue that this is no different than states having different laws pertaining to a lot of things, such property. That is a deceptive analogy. Property laws are based on centuries of experience with property, as well thousands of cases and judges who understand those cases. Conversely, these new AI state laws are vague, untested, and written in ways even judges might not understand. With AI, there is no playbook to follow. It is simply a guessing game with huge ramifications.
What is Really Going On Here
This is classic rent-seeking, politicians making laws that look good on paper but end up helping big companies which have the big money necessary to influence legislation to their ends. Rather than admit that all of this red tape gives big tech a competitive advantage, lawmakers play up the emotional angle, with “think of the children” style arguments. This makes it hard for smaller companies to oppose the legislation, or even ask questions, about these onerous laws that only big companies can afford to follow.
The result? Less innovation, fewer startups, and more wasted money.
A Better Way Forward
Instead of 50 different sets of rules, we need:
One clear national standard
Coordination between government and international organizations
Laws that solve real problems without creating new ones
And we need lawmakers who understand the cost of getting this wrong.
Final Thoughts
If we continue down this path, we will strangle the very innovation these state laws claim to protect. If you want to protect the public, fine. But do it in a way that does not kill small business in the process.
Let’s pass smart, federal AI laws, not a patchwork of state AI laws that make it impossible for anyone but the biggest players to compete.
If you are an inventor, entrepreneur, or startup with a technical idea, the good news is that a recent shift in United States Patent and Trademark Office (USPTO) policy limits how third-parties can attack your patents. The even better news is that this new policy shift can make a patent drafted today stronger than one drafted under the old policy.
This rare “window of opportunity” at the USPTO gives inventors wanting to file new patent applications some new advantages, and removes some old disadvantages, in the patent application process.
What Changed: USPTO Restricts Use of Applicant Admitted Prior Art in IPRs
What does that all mean? Basically, if a challenger wants to try to invalidate your patent, they typically go through a process known as an Inter Partes Review (IPR). An IPR allows a patent challenger to argue to the Patent Trial and Appeal Board (PTAB) that your patent is invalid. The challenger typically argues that the Patent Office did not take into account certain known technologies (prior art), or that the Patent Office failed to combine various pieces of prior art to show that your invention was obvious, and therefore unpatentable.
Prior to September 1, 2025, if a challenger could not find all the elements of your patent claims in the prior art, the Patent Office allowed the challenger to instead use “general knowledge” to fill in the gaps to show the invention was obvious at the time the patent application was filed. “General knowledge” includes things like expert testimony, common sense, and applicant admitted prior art (AAPA) submitted as part of the patent application.
As of September 1, 2025, however, challengers are no longer allowed to rely on this type of general knowledge to fill in the missing elements to invalidate your patent. From now on, a challenger must specify exactly where in the prior art that each and element of your patent claim can be found. If they cannot find each and every element in the prior art, their IPR petition must be rejected by the PTAB for failing to meet the statutory requirements of 37 C.F.R. § 42.104(b)(4).
Why This Matters for You (the Inventor)
Stronger defense: Now you no longer have to choose between adding AAPA to support your patent application and avoiding having that AAPA later used against you in an IPR.
Strategic drafting matters more than ever: The way you frame your specification, disclosures, and claim dependencies against the prior art can make a difference on the defensibility or your patent under the new standard.
Limited Window: As the Patent Office can change its mind at any time, if you have been considering filing for a patent to protect your invention, filing during this favorable window of opportunity could mean the difference between your patent being valid or invalid.
A battle avoided is always better than a battle won: Now that challengers will have a more difficult time finding prior art for every element of your patent claims, they will be less likely to file an IPR in the first place, making your patent more valuable and less likely to be challenged.
Other Patent Trends Worth Watching (and Using to Your Advantage)
To position your patent well, here are trends you should leverage in 2025:
Trend
Why It Helps You
Action Tip
AI and cross?discipline inventions
AI is fueling invention across biotech, materials, software, energy.
Frame your invention to highlight cross?field novelty; look for overlap points
Quality over quantity
Patent portfolios are judged on impact, not sheer number.
Focus on broad claims with commercial strength; avoid weak “kitchen sink” claims
Backlog & examination delay
USPTO has a record high backlog of unexamined applications.
Determine your optimal timeline for an issued patent and leverage priority examination if necessary; plan your timelines realistically
Global patent competition
China (and others) are dominating patent filings globally.
Don’t ignore foreign patent strategy; international filing makes a difference
How to Make the Most of This Opportunity
Talk to a patent attorney you trust and who knows your technology: make sure your patent application fully describes your invention and has claims tailored to take advantage of new Patent Office policies.
Avoid overly broad admissions in your patent specification: focus on differentiating your invention from the prior art.
Build flexibility into claims: structure your patent specification and claims so that fallback positions remain meaningful.
Plan for global filings: assess countries where your invention may have commercial value and pursue international protection accordingly.
In a major shift that could shape the future of artificial intelligence innovation, U.S. Patent and Trademark Office (USPTO) Director Kathi Vidal has been joined by newly appointed agency director John Squires in taking a strong stand in support of patent protection for AI technologies.
Squires recently made headlines for pushing back against a Patent Trial and Appeal Board (PTAB) decision that, if allowed to stand, could have blocked patent protection for wide swaths of AI innovation. Squires warned that the PTAB’s logic risked making “much of AI unpatentable,” a conclusion he called legally flawed and dangerous for American competitiveness.
The message is clear: the USPTO is not only listening to inventors but also working to clear the path for innovators using artificial intelligence. And it could not have come at a better time.
AI Innovators Have Been in Limbo
In recent years, developers of artificial intelligence have faced increasing uncertainty about whether their inventions qualify for patent protection. With courts and tribunals tightening rules around what counts as patentable subject matter, many AI-related patent applications have hit roadblocks. The concern has been that a luddite patent examiner might view an invention applying AI to solve a problem as a patent-ineligible “abstract idea,” especially if the solution involves software or data.
In July, the USPTO issued updated guidance clarifying that AI inventions can be patentable if they are tied to specific, practical outcomes and do more than simply apply AI to an existing process[1]. This was a welcome step, but the recent comments from Squires go further. They show the Patent Office is willing to push back against internal decisions that don’t align with the goal of protecting American innovation.
A Green Light for AI Patents?
Squires’ remarks are more than just legal commentary. They signal a green light for AI inventors who may have been holding back on filing patent applications due to fear of rejection. They also provide patent attorneys with new ammunition to fight back against overly narrow interpretations of patent law.
The key takeaway for inventors and startups working on artificial intelligence is this: the USPTO is not your enemy. In fact, its top leadership is actively working to ensure that AI innovation gets the legal protection it deserves.
What Should You Do Next?
If you’ve been developing AI technology, now is the time to take action. Review your innovations to see what may be patentable. Focus on outcomes. If your invention produces a real-world result, not just a calculation or prediction, it may very well qualify for a patent under the current guidance.
But not all patents are created equal. If you hired 50 different patent attorneys to draft a patent on your new AI system you would get back 50 different patents. Each of these patents would all have a different likelihood of being granted and all have different scopes of protection. To give yourself the best chance of getting your AI patent issued with the broadest scope of protection, your patent claims (the legal scope of your patent) must focus on specific, technical improvements, rather than current AI buzzwords. As always, the devil is in the details. You cannot get a great AI patent unless you start with a great AI patent application.
And don’t wait. While the USPTO has extended its comment period on AI-related patent eligibility to October 16, 2024[1], there’s no guarantee these new interpretations will last forever. Once public comments are reviewed and finalized, the Patent Office may adjust its approach. Filing early gives you the best chance of locking in your rights under conditions more favorable than they have ever been, and may ever be again.
Final Thoughts
For those of us who have been working with inventors navigating the muddy waters of AI patent law, Director Squires’ comments are a breath of fresh air. They confirm what we’ve long argued: AI is not some magical concept outside the reach of patent law. It is a tool, one that, like any other, can be used in innovative ways that deserve protection.
The Patent Office is finally catching up to the technology. And for once, it’s good news.
In September 2025, Popeyes Louisiana Kitchen, LLC filed a federal lawsuit in Iowa against former franchisee Asif Poonja and Jam Equities, et al, alleging unauthorized continued use of Popeyes’ branding after termination of their franchise agreements. Although the five restaurants in this case are all located in Iowa (Des Moines, Coralville, Cedar Rapids, Waterloo, and Dubuque), Popeyes filed this case in Florida federal court pursuant to the venue selection clauses in the agreements the defendants signed with Popeyes.
Below I break down the key trademark issues raised in the suit, and what businesses with licensing or brand aspirations should take away from this dispute.
Facts Alleged in the Lawsuit
Popeyes terminated the franchise licenses for five Iowa locations on August 29, 2025, citing the defendants’ failure to meet food-safety and brand standards.
After termination, the defendant locations allegedly continued operating using Popeyes marks — signs, menus, branding, uniforms, etc.
Popeyes seeks injunctive relief (to force the removal of branding), monetary damages, and a non-compete restriction (cannot operate a “chicken restaurant” within 10 miles of the current locations for two years).
Popeyes claims that the continued use of its trademarks “risks serious, immediate, and irreparable harm” and causes brand dilution and consumer confusion.
From that foundation, we can see how multiple trademark principles are being marshaled in this fight.
Trademark Issues at Play
1. Unauthorized Use After License Termination
A core claim in this case is trademark infringement under the federal Lanham Act. Once the franchise relationship ends, the former franchisee (or operator) no longer has authorization to use the franchisor’s trademarks. Continued use after license termination is classic trademark infringement, unless some residual rights survive (which is rare and would need to be contractually stipulated).
To prove trademark infringement, Popeyes must show:
The trademarks are valid and protectable
They own the trademarks
The defendant used the trademarks (or confusingly similar marks)
The use was “in commerce”
The use is likely to cause confusion (or dilution) among consumers
As the defendants continue to operate under the Popeyes name and use the branding after termination of the trademark license in the franchise agreement, it appears Popeyes will be able to provide evidence of each of these factors.
2. Consumer Confusion
Popeyes argues that continuing to operate these restaurants as “Popeyes” is likely to create confusion, namely, consumers are likely to believe they are visiting a genuine, authorized Popeyes, held to corporate standards. This likelihood of confusion may harm the brand’s reputation if the restaurants underperform, violate safety or quality norms, or diverge from standardized operations. Even if the quality is high, the defendants are still profiting off of the goodwill Popeyes has built in their trademarks.
Because trademarks are tied to guaranteeing a certain quality and source, unauthorized misuse can damage the brand, even if consumers don’t consciously think “this is fake.” This is another form of trademark infringement.
3. Trademark Licensing, Control, and Quality Standards
One of the canonical rules in U.S. trademark law is that a trademark licensor must maintain control over the licensee’s quality to avoid abandonment or invalidity. In a franchise context, that means contractually enforcing operations standards, inspections, branding rules, etc. If Popeyes can show it monitored and enforced those standards (and terminated when breached), it helps show the marks’ integrity hasn’t been diluted.
The fact pattern here suggests Popeyes claimed these franchisees repeatedly failed inspections and brand compliance. That supports the notion that Popeyes was actively policing operations. In cases like this, Popeyes does not have the option of turning a blind eye to ongoing trademark infringement. To do so would risk the Popeyes trademark falling into the public domain. This is exactly what happened to previously valuable trademarks like aspirin, escalator, yo-yo, etc.
4. Injunctive Relief and Irreparable Harm
Trademark owners often seek injunctions, not only damages. An injunction is an order from a court that one party must do, or must stop doing, something. In this case, the injunction would prevent the defendants from using any of the Popeyes trademarks. Popeyes’ assertion of “irreparable harm” is textbook, as quantifying ongoing brand reputation damage is a difficult task. The sooner the unauthorized usage is enjoined, the less risk of lasting brand erosion.
The defendants may (if they contest) argue that monetary damages are sufficient or that an injunction is overbroad. If the court grants an injunction, the court may require Popeyes to post a monetary bond with the court that would go to the defendants if it is later determined, based on new evidence, that the injunction was not warranted.
An injunction may also apply to the non-compete / exclusion zone requested (10 miles, two years). A court may find the metes and bounds of this non-compete to be overly aggressive may enjoin enforcement of the exclusion or may “blue pencil” in more appropriate numbers given all of the other factors at play.
Many franchise agreements include clauses requiring de-identification after termination: the franchisee must remove all signage, logos, uniforms, and references to the brand. Popeyes references that the defendants failed to deidentify. That is a contractual claim (breach of contract) that runs in parallel to the trademark claim.
Because the use post-termination is unauthorized both by trademark law and by contract, Popeyes can layer multiple causes of action (infringement, unfair competition, contract breach) in this one case.
Popeyes is seeking monetary damages, likely including:
Past profits made unlawfully
Lost licensing fees
Possible trebled damages if willful
Attorney’s fees (if the case qualifies)
One big benefit of having your trademark federally registered is that availability or treble damages and attorney’s fees in cases where the defendant willfully continues to infringe the trademark. The availability of injunctive relief (stop usage, require removal of marks) is also critical to stemming the accumulation of trademark damage. The non-compete / exclusion zone is a more novel remedy in some cases, though not unheard of in franchise litigation, but its enforceability depends on reasonableness, statutory and public policy constraints in the relevant jurisdiction.
7. Other Possible Remedies: Trademark Dilution and Tarnishment
Even if the defendants are using the Popeyes trademark in association with goods or services that Popeyes does not offer, Popeyes could argue dilution: that the value, distinctiveness, or strength of the POPEYES® marks is weakened by unauthorized use on other goods or services. The continued use of the mark without control could “lessen the capacity” of the famous mark, in this case Popeyes, to signal a consistent source.
If the allegedly non-conforming restaurants have health or cleanliness issues (as Popeyes cites), Popeyes could make an argument for tarnishment: that the defendants are associating the Popeyes trademark with bad conduct, diminishing the value of the trademark in the eyes of consumers.
Strategic & Practical Lessons for Brand Owners
From the Popeyes case, several “lessons learned” emerge, especially for companies licensing trademarks, franchising, or otherwise extending their brand reach.
Have clear contract language about post-termination obligations. Include robust deidentification, signage removal, and transition procedures. Explicitly state that all rights to use the trademarks cease at termination.
Federally Register All of Your Trademarks and Logos
Federal trademark registration offers many advantages, from proof of ownership, to national rights, to access to federal courts, to the possibility of treble damages and attorney’s fees.
Retain tight quality control. A trademark license without actual quality control invites a court finding abandonment or incontrollability. Inspections, audits, standards, and especially enforcement, all help.
Act swiftly on breaches. The stronger your position to show you promptly objected and terminated non-compliant franchisees, the more credible your claim of controlling dilution.
Venue selection clauses in all of your agreements are important.
If Popeyes had not included a Florida forum selection clause in its franchise agreements, it would likely have to try this case in Des Moines, Iowa, increasing both costs and attorney time.
Be careful with non-compete or radius restrictions. These can provoke fairness or antitrust objections. They should be proportional and justified by protecting the brand network, not punishment.
Monitor post-termination compliance aggressively. Even after termination, monitor whether the former licensee is persisting in use, and consider legal action earlier to prevent brand damage.
Document everything. Inspection reports, notices of breach, correspondence, logos used, deidentification communications — these build your case.
Consider injunctive relief early. The longer unauthorized use continues, the more harm accrues (harder to reverse). Courts often expect brand owners to ask early, not wait until the damage is entrenched.
Possible Defenses & Challenges
From the defendant’s perspective, the potential counterarguments, although tenuous, might include:
Lack of confusion. They might argue that consumers would know the restaurants are no longer corporate, or that disclaimers mitigate any confusion.
Overbreadth of injunction or non-compete. They may challenge the 10-mile, 2-year limitation as unreasonable or oppressive.
Contractual ambiguity or waiver. They might argue that post-termination rights or residual use (if any) were allowed or that Popeyes waived enforcement.
Statutory limitations or public policy. They may invoke state law limitations on restrictive covenants or competitive restraints.
Why This Lawsuit Matters to IP & Brand Owners
This is a clear high-stakes example of how a major brand protects its marks in the franchise context. While the damages in this particular case may be small, failure to stop the trademark infringement could lead to these multi-million-dollar trademarks entering the public domain.
It reinforces that termination is not the end of the story — trademark enforcement must continue.
It underscores that brand protection is not just about registration; it is about policing and asserting rights consistently.
It provides a roadmap (positive and negative lessons) for companies structuring licensing and franchise systems going forward.
Conclusion and Call to Action
The Popeyes litigation is more than just a messy franchise dispute, it is a vivid case study in why trademarks require ongoing vigilance, contractual clarity, and swift enforcement. For brands, the stakes are real: permitting unauthorized use can steadily erode the goodwill you’ve painstakingly built until it is completely gone and you have no trademark left to enforce.
If you run a franchise system, license your brand, or plan to expand via partnerships, now is the time to audit your agreements, enforcement policies, and monitoring systems. You need a consistent strategy for licensing and enforcing your trademarks. Although the prospect of diving into the legalities of trademark protection may seem daunting, it is infinitely easier to develop this strategy now, than when your trademark issues end up in federal court.
I am a patent attorney, but the posts on this site are not legal advice. Peculiarities of the law, the facts, the client and the lawyer (mostly), make every case unique. If you have a question about a particular case, contact me directly.
No information you send me or post to this site is confidential. By posting or sending anything to me, you authorize me to use it for any lawful purpose.
If you have questions about Internet law, you may purchase my book Cyber Law. You may also visit my Iowa Patent Lawyer website or stop by our offices at 516 Walnut, Des Moines, Iowa.
Since 2003
Recent Comments