Hello again everyone.  I suspect a number of you out there have been wondering when I was going to get around to today’s topic.  Hell, even I was wondering about it.  Back in January, I identified ChatGPT as one of my top ten things for in-house lawyers to pay attention to in 2023, but I must admit that I am shocked at how quickly it has overwhelmed the debate about legal tech.  To be honest, I was hoping for a moment when the pace of “new” ChatGPT information would slow down.  It hasn’t.  And likely won’t.  But I think it’s important enough of a topic to at least try to take a snapshot of where things stand today for in-house lawyers when it comes to ChatGPT (or more generically, generative AI).  Back when I was general counsel, I would have grabbed a few folks on my team and said, “Let’s jump into a conference room and whiteboard ten things we need to know about ChatGPT.”[1]  As I have been thinking about and using ChatGPT, I realize we may need two whiteboards – there is a lot to cover!  What I have done is read a lot about it and I have tried it out a number of times with many different asks.  All to put myself back in the chair of an in-house lawyer and cut through the clutter to make sense of just what this tool is and what can it do to help me and the legal department.  So, get your dry markers out, grab a big mug of coffee, and join me in the conference room as this edition of “Ten Things” tries to set out what in-house lawyers need to know about ChatGPT (and be sure to click through the links and check the footnotes for – a lot of – extra information):


1.  What is ChatGPT?  I tried to think of a cool answer to this question, but I gave up.  ChatGPT is just a chatbot.  A really powerful chatbot though, one that allows for incredibly humanlike conversations and responses to questions.  Think of it like a useful version of the old Microsoft “Clippy” character who was universally hated and despised by all that came across its leering mug and unhelpful answers.[2]   But, I digress.  Beyond the simple “it’s a chatbot” lies a far more interesting definition.  ChatGPT is a conversational AI system that listens, learns, and even challenges you at times.  It is powered by a large-scale language model (“LLM”) trained by a company called OpenAI on a massive amount of textual data (but, it is not – as of the date of my writing this – connected to the Internet in real time).  It interacts with you in a dialogue format (not a list of links), which allows you to ask follow-up questions – and cut and paste the text into your own work product).  It will admit mistakes, challenge incorrect premises, and reject inappropriate requests (trust me, I ask it a few of those!).  It will also make shit up.  Which is kind of a negative unless you are a politician.[3]  While I am focusing on ChatGPT, there are many different flavors of generative AI tools out there.  Here are just a few that in-house lawyers should check out:[4]

  • ChatGPT – the most commonly used and free version of the tool.  Uses GPT 3.5 as its engine (which is a fairly recent improvement over GPT-3, the engine that started it all in late 2022).
  • ChatGPT+ – a far more powerful version of ChatGPT available for $20.00 per month.  Uses GPT-4 as its engine which is vastly superior to GPT-3.5 in terms of power, learning, and output.  This is the version I would use as an in-house lawyer.
  • Bard – Google’s version of ChatGPT.  Currently available only in the US and UK.  It’s a piece of junk at the moment, but it will get better.
  • Bing Chat – Microsoft’s version of ChatGPT that utilizes GPT-4 as its engine.  Easy to use.  This could make Bing relevant (words I never thought I would write…).
  • Microsoft Copilot GPT-4 for Office and Word – Microsoft is also (smartly) integrating generative AI into its Office suite of tools, the tools many, if not most, in-house lawyers already use daily, in particular Word.  Currently being tested, and likely rolled out widely later this year.  Watch for this one – it will be a game-changer for in-house lawyers in my opinion.
  • LawDroid – a version of generative AI aimed specifically at lawyers.  Not widely available yet, but you can sign up for early access.
  • Spellbook – another GPT4 tool trained on legal language content and aimed squarely at lawyers.  Alas, there is a waitlist for this one too.
  • Co-Counsel – an “available now” AI tool from Casetext and legal-focused.  You can get a free trial but then you’ll need to pay.

There are certainly more generative AI tools out there for lawyers.  My advice is to start with the basics (free or low-cost) and keep your eyes open for tools that better fit the world in-house lawyers live in (and do a better job of avoiding the ChatGPT pitfalls, which I outline below).

2.  Will it steal my job?  Highly unlikely. It may change it – a lot.  But, in these still very early days, generative AI will not replace lawyers.  In fact, it needs humans to stay involved, especially when used for legal work.  To start, the bar will not allow AI to replace us; the practice of law will require humans in some capacity no matter what.  Second, you must validate everything it spits out, another reason why the bar won’t allow AI to replace lawyers (see below for more on the problems with ChatGPT and ethical obligations).  Third, ChatGPT does not understand context, nor can it discern whether it is being used to come up with the answer the user wants vs. the correct legal answer (i.e., right vs. wrong, but also, as Charles Spurgeon said, right vs. almost right).[5]  Only people can do that.  Generative AI can only spit out data, it cannot engage in critical thinking.  Fourth, when it comes to legal work, most clients will want to talk (or be able to talk) to a person, not a chatbot (e.g., the CEO is not going to type questions into a chatbot to get advice about a merger or a new regulation – they want their legal team at the table).  And fifth, the day that all that I (and you) do can be reduced to a list of mindless tasks that can be automated is not a day I see showing up on my Outlook calendar in my lifetime.  The better news, however, is that, as a tool, ChatGPT should – eventually – make your life easier and allow legal departments to increase efficiency without adding headcount (or cutting headcount) and without having to invest large sums of money.  ChatGPT/generative AI will be another tool you can use to streamline tasks and reduce the amount of mundane work you must deal with (for example a chatbot to answer frequently asked questions) and, instead, free you to work on complex, high-value projects where you truly show your value.  Generative AI will also create new jobs, including the need for lawyers who specialize in preparing prompts and training the system for the needs of a particular legal department at a particular company.[6]  Still, nothing is 100% guaranteed, so don’t rest on your laurels here, it’s also good to be a little afraid that you might be replaced – even with a machine.  Don’t be a dummy. Always work to make yourself indispensable.

3.  How does ChatGPT work? Several years ago I wrote a blog post on artificial intelligence.  It was high on hope, but not much in terms of truly useful “here and now.”  ChatGPT will change that and bring AI tools for in-house lawyers to legal departments for use daily, seamlessly, and effortlessly[7] – as a natural extension of the legal department.  ChatGPT builds on long-standing AI principles that originated in the 1950s.  At its core, it relies on machine learning, i.e., machines (computers) programmed to mimic human learning.  Specifically, it works by attempting to understand your question (“prompt”) and then start setting out strings of words that it “predicts” will provide the best answer, all based on – and currently limited to – the data set it was “trained” on.  In other words, it scans a shitload of information and writes the best answer possible based on the data it has access to.  Just like you do as a lawyer, only much slower – but with innate human skills that AI cannot yet (if ever) duplicate. The gigantic data set is something called “tokens” which I will not pretend to understand.  But tokens were used to help ChatGPT form a “deep learning neural network” – which is basically a super complex computer algorithm modeled after the human brain.  This allows ChatGPT to “learn” patterns and relationships in the data set and then create humanlike responses by predicting what text should come next in any given sentence.  Okay, that’s way more than you probably need to know (and way more than I want to write about this point, so we’re moving on – but feel free to ask ChatGPT how it works or Google it if you want more of the details).

4.  How do I use it?  Finally – time for the fun part!  Actually, this is a much harder question than it may seem.  Currently, you utilize ChatGPT via what are called “prompts.”  Prompts are questions, instructions, or requests that you type into the tool to trigger the process.  The complexity comes into play when you want to move beyond a simple request and ask ChatGPT multiple questions, make multiple requests as part of one prompt, or refine the ChatGPT results by refining the prompt or adding to the data ChatGPT is considering.  The good news is that you can do all of this.  The “bad” news is that it takes practice and time.  The latter is something most in-house lawyers don’t have a lot of.  So, you need to balance the potential gains from utilizing generative AI with the cost in terms of time spent and, as discussed later, sometimes crappy results.  For purposes of today, I am going to keep things simple (but give you some resources to go deeper or enhance how you might use ChatGPT in the legal department).  In short, treat it like a young associate.  You must brief it, give additional information, ask clarifying questions, and likely fix the work product.  But, to start, here are a few simple prompts I have created or found that can be easy ways to get started with ChatGPT.  Once you have access to ChatGPT (or whatever generative AI tool you wish to use), try playing with the following prompts:[8]

  • What is the standard for [set out legal issue] in [x] jurisdiction?
  • Outline the steps needed to do [y]
  • Create a checklist for…
  • Draft an email explaining [scenario]
  • Draft a contract for [scenario]
  • Draft a contract clause for [scenario]
  • Set out the pros and cons of [x]
  • Prepare a presentation from the legal department to the business on [topic]
  • Set out ten things I need to know about [topic]
  • Produce five titles for [topic]
  • Explain the [legal topic]
  • Summarize this meeting transcript
  • Summarize this agreement and identify the five most important terms
  • Prepare a term sheet for [name] type of deal containing these key terms [list]
  • Write [x] in the style of a business person and not like a lawyer
  • What are the best law firms for [issue] in [jurisdiction]?

While playing with ChatGPT and thinking through uses for in-house lawyers, I came upon a few great articles from one of my favorite blogs, Contract Nerds, in particular, the “Chatting Contracts with GPT” column by Laura Greenberg.  These are worth reading because they show specific prompts for incredibly useful ways to use ChatGPT for contract drafting, i.e., how to use ChatGPT to draft explanatory comments in contracts and how can ChatGPT help us negotiate contracts more efficiently.  Lastly, check out this excellent article from Slaughter & May, Generative AI: Practical Suggestions for Legal Teams.

5.  What can it do for me today?  This is the part that has truly blown me away, i.e., the incredibly long list of things that ChatGPT/generative AI can do for in-house lawyers right now.  I was tempted to just list a handful, but I think it’s more interesting to just go ahead and set out a fulsome list (just know that my list only scratches the surface).  If not involving ChatGPT alone, I have included links to different generative AI tools that can do these tasks today (in addition to the AI tools I set out above in No. 1).  Lastly, I’ll note that I cannot yet vouch for most of these, but the sources I have used are very reliable:

6.  Let’s use it!  I know many of you have already used ChatGPT but a lot of you have not.  So, let’s go through the steps of a simple problem and see what we get.  To start, create an account on OpenAI for the free version of ChatGPT and then get to the opening screen (sorry that the screen capture quality is a bit fuzzy):



Next, type your request in the text box (“Send a message…”).  Let’s try “Explain the attorney-client privilege in an email to a non-lawyer.”  Then press enter.  Here’s what I got:

Subject: Understanding Attorney-Client Privilege: Your Legal Communications Remain Confidential

Dear [Recipient’s Name],

I hope this email finds you well. I wanted to take a moment to explain the concept of attorney-client privilege, an essential element of the legal profession that ensures confidentiality and promotes open communication between attorneys and their clients.

Attorney-client privilege is a legal principle that protects the privacy of communications exchanged between a client and their attorney. It establishes a sacred trust, allowing clients to share sensitive and confidential information with their legal counsel without fear of it being disclosed to others, including in a court of law. This privilege is considered fundamental to the effective functioning of the legal system and is upheld in most jurisdictions.

Here are a few key points to help you understand the attorney-client privilege better:

Confidentiality: Attorney-client privilege establishes a duty of confidentiality, requiring attorneys to keep their clients’ information strictly confidential. This obligation extends beyond the duration of the attorney-client relationship and even survives the client’s passing.

Scope of Protection: The privilege covers any form of communication between an attorney and their client, including in-person discussions, phone calls, emails, text messages, and written correspondence. It applies to both civil and criminal matters and protects not only what is explicitly shared but also information that is implied or understood within the context of the attorney-client relationship.

Limited Exceptions: While attorney-client privilege is robust, it is not absolute. There are some exceptions that may allow disclosure of otherwise confidential information. For example, if a client intends to commit a crime or harm someone, an attorney may have a legal duty to report it. Additionally, if the client waives the privilege or if a court orders the disclosure of certain information, the attorney may be required to comply.

Legal Advice and Representation: Attorney-client privilege covers communications made for the purpose of seeking legal advice or representation. This means that clients can be open and honest with their attorneys about their situation, concerns, and any potentially incriminating or sensitive information, without fear that it will be shared with others.

It is important to note that attorney-client privilege only applies when there is a formal attorney-client relationship in place. Simply seeking casual legal advice or discussing legal matters with someone who is not your attorney does not establish the privilege.

Please remember that I am not providing legal advice specific to your situation through this email. If you require legal advice, I encourage you to consult with an attorney to discuss your circumstances in detail.

I hope this explanation helps clarify the concept of attorney-client privilege for you. If you have any further questions or require additional information, please feel free to reach out.

Best regards,

[Your Name]

As you can see.  It’s a pretty good summary in email form that I can easily edit and put into my own voice or customize it for my company’s particular circumstances.  Even better, it only took a few seconds for ChatGPT to prepare this – saving me the 10 or 15 minutes it would probably take me to put it together (which multiplied over the course of a month is a lot of time saved).  And, because the topic was general, I do not have to worry about confidentiality issues.  While I will not set these out in this post, while I was playing with it, I next asked it to “Prepare a simple mutual non-disclosure agreement for a business using Texas law and exclusion jurisdiction in Texas” and got back a usable first draft.  Then I asked for a mutual indemnity clause.  Same result, a usable first draft produced in a few seconds.  It won’t replace me (for a long while anyway, if ever) but it clearly can be a very useful tool when used properly.

7.  Ethics issues.  First, let’s cut to the chase: ChatGPT has no ethics.  Seriously, it’s just a machine.  It has no ability to discern, apply context, recognize when it is making things up, or deal with or express emotion.  It’s just a (potentially) really useful tool.  Always keep this top of mind.  Second, under the rules of professional responsibility, you will have many ethical obligations to watch out for and comply with when using generative AI.[10]  Here are some of the most important under the ABA Model Rules:

  • Rule 1.1 – the duty of technical competence for lawyers (basically, all lawyers must stay up to date on technological developments that impact the practice of law).
  • Rule 1.4 – communications (specifically Rule 1.4(a)(2) and the need to inform the client that you will be using AI to assist with providing your services).
  • Rule 1.6 – the duty of confidentiality (when using generative AI tools, you must ensure that any client information you enter is not confidential or, if it is, that the tool you are using will protect that confidentiality).
  • Rule 5.1 – the duty to supervise (in particular, for those you supervise you must ensure they know the applicable ethical rules and comply).
  • Rule 5.3 – the duty to supervise non-lawyers (lawyers cannot outsource their work to non-lawyers, like ChatGPT.  They must stay involved).

Several others may apply as well (see, e.g., ChatGPT: A Lawyer’s Friend or Ethical Time Bomb? A Look at Professional Responsibility in the Age of AI).  Similarly, while unlikely, in-house lawyers are not immune from malpractice lawsuits.  Meaning, if you use generative AI for legal work, you must know what it is doing, you must review the work product, and you must not take anything you get from ChatGPT at face value.[11]

8.  What are the risks?  While it seems like using ChatGPT is a great idea, in-house counsel must go into it with their eyes wide open.  There are many problems with these tools currently (and likely for a long time).  Here is a partial list of issues I have come across so far:

  • Limits on information – currently, ChatGPT only uses information made available to it through September 2021 and it does not connect to the internet when searching.  That’s a problem (unless you don’t care about the last two years).  I understand that an update is underway, however.
  • Memory Limits – currently, ChatGPT-3.5 can accept about 3,000 words max.  ChatGPT-4 can accept about 12,300 words max.  That is good, but not enough to do hard-core legal work, especially around analyzing briefs and motions.[12]  Plus, once you cross over the word limit, ChatGPT “forgets” everything you entered before that.  Ugh.
  • It makes stuff up – ChatGPT sometimes makes up answers.  These are called “hallucinations” (which is not something a lawyer wants to hear about its assistant, real or virtual).  It has also been described as overconfident and unaccountable.  Not a great combination either.  This is why it is critical that lawyers double-check any material they get from generative AI tools, especially if they are using it for any type of analysis or legal writing.
  • “Garbage in/garbage out” – everything looks great typed up, but unless you are providing the database to train the tool, you must acknowledge that there is a lot of garbage out there and ChatGPT is sifting through the garbage heap to come up with answers.
  • IP infringement – ChatGPT does not care what sources of information it uses or whether that information is protected by copyright or trademark law.  It will give you whatever it thinks best answers your prompt (or make shit up).  A big question is whether generative AI tools can ingest “everything” and call it “fair use” under US copyright laws. If not, what does that mean for the future of these tools (and the cost)? There are also, generally, no citations or footnotes given to attribute source material (though you can ask for them – and even then, as I have found out, those can be 100% bullshit).  In other words, it is a plagiarism timebomb from ACME products waiting to explode on the unsuspecting coyote, i.e., you.  Unless you are using the tool in a closed loop of information, IP infringement is a real risk.
  • No confidentiality – unlike Las Vegas, what happens with your ChatGPT prompt doesn’t always stay within your prompt and whatever information you enter into ChatGPT can find its way into the public domain.  Some Samsung employees found this out the hard way.  Fixes are underway, but until that happens, be careful how you word your prompts and use dummy or fake names.  And until this is fixed, the tool will not be used at an enterprise level.
  • Offensive content – even if it is limited to data through September 2021, all of the ugliness of the internet is part of ChatGPT’s learning, meaning ugliness can – and has – come out in responses.  It can be as toxic and hateful as, well… just look around you.  Watch out for this along with defamation issues if you are going to use ChatGPT material in something you will make available to others.
  • Zero accountability – generative AI is not rational, has no ethical boundaries, and doesn’t care what sources it uses (or if it must make up the answer).  In other words, there is zero accountability.  That’s okay, so long as you know that you are accountable.
  • Privacy laws – personally identifiable information you put into ChatGPT may be protected by various privacy laws, e.g., CCPA, GDPR, and others.  If so, how does ChatGPT allow you to comply?  How do individuals exercise their right to delete/be forgotten, correct or amend, and get copies?[13] Until this gets worked out (and it will), it’s best to use fake names and broad inquiries as much as possible. And, keep in mind that generative AI can be used as part of a ransomware or other cyber-risk scheme.[14]  Yep, the fun is endless!
  • Lack of policies/regulation –  many companies (and legal departments) have been caught off-guard by ChatGPT and its popularity.  We will need policies and procedures put in place to protect the business and ensure that employees are using the tools properly (if at all).  Likewise, regulators are sprinting (lurching?) to enact legislation that covers generative AI.[15]  It is the wild west out there at the moment.  I believe regulation is needed, but my hope is that such regulation doesn’t go so far as to kill a promising tool in the cradle.  Overall, generative AI is definitely an area where the legal department can show value as the company watchdog.

While these are all pretty scary it does not mean that you should just ignore ChatGPT and hope it goes away.  It won’t.  So, like it or not, welcome to the brave new world and keep your guard up along with your willingness to answer questions and solve problems – because that’s what lawyers do.[16]

9.  Resources.  As I mentioned in the introduction, I have included a lot of links and footnotes in this post to help you go deeper into particular points.  There is so much material out there, I thought it would be helpful to set out a few articles or resources I found that are particularly helpful for in-house lawyers just getting started with ChatGPT/generative AI:

10.  What do I do next?  In my experience, lawyers are generally slow to adopt new technology.  We are naturally skeptical and generally see the problems with something new vs. the benefits.  I get that, but given what I said above, ignoring generative AI is not an option.  So, as we finish our whiteboard session, here is what I would be telling my team the legal department needs to do next:

  • Embrace it.  It’s here to stay.  But act with restraint and caution. Verify everything, trust nothing.[18]
  • Develop legal department and company policies regarding the use of ChatGPT (and note the down sides of all-out bans).
  • Start small and with free products and low-risk tasks (understanding the risks) and get your feet wet. Then move to ChatGPT+ (the more powerful paid version).  Finally, look for established companies offering generative AI products and use those to truly establish a foundation for AI use in the legal department, e.g., Office365, Google, Thomson Reuters, Lexis Nexis, etc. But keep an eye out for up and comers as many companies are starting to tailor AI tools for in-house legal departments.
  • Subscribe to the AI Exchange if you are serious about using AI tools.[19]
  • Do it as a team, i.e., figure out how best to make ChatGPT work for everyone in the department.  Consider appointing someone (or asking for a volunteer) to become the ChatGPT guru on the legal team.
  • Keep data privacy concerns top of mind.
  • Learn how to draft prompts that work for in-house legal research and needs.[20]
  • Stay up to date.  Things are happening fast, and the ground keeps shifting.  Keep looking for current information about generative AI from trusted sources.
  • Watch out for regulatory developments (and if you see developments that might hurt or help your company, get involved in the discussion).[21]
  • Understand your state’s ethical obligations around the use of ChatGPT.
  • Find a way to make the department’s AI tools available to the business as part of a self-help initiative, e.g., FAQ chatbot or document assembly tool on your intranet site.  Just be sure it doesn’t suck.  Then tout it as a way the legal department is innovative and cutting edge.  Take credit for it!
  • Take the lead (or join the team) looking at ChatGPT for use by the company as a whole.  The legal team should be front and center here and leading from the front.


I have to say, this was one of the hardest yet most fun “Ten Things” posts I have ever written.  I hope that this provides a good starting point for in-house lawyers looking to get on with or enhance their use of ChatGPT and generative AI tools.  When I came out of law school, we were using typewriters and dictaphones to draft documents and poring through books and legal compendiums for research.  When I take a moment to look back and then think about where we are today with technology, it’s quite stunning (though I still prefer books for legal research).  I probably cannot even imagine what all of you will see over the next thirty years, but I do know that it will be exciting and amazing.  It will also be packed with unintended consequences, some good and so bad – really bad.  On that front, I do believe that it will take a long time for ChatGPT to live up to all the hype when it comes to legal work.  Yet, I have no doubt it will get there.  This is truly one of those game-changing moments in history, especially in regard to the practice of law. And you have a front row seat and a part to play.  As in-house lawyers now is not the time to shy away from generative AI, it’s time to run to the fire.  When you get there, be smart and thoughtful about how you, the department, and the company use these tools.  Do your research, be practical and open to new ideas, and give your best advice at all times.  You know, be a lawyer.

Sterling Miller

May 31, 2023

Book six is underway, but my fifth book, Showing the Value of the Legal Department: More Than Just a Cost Center is available right now, including as an eBook!  The first printing sold out, but the second printing is now back in stock.  ABA robots are standing by to take your order!  You can buy it HERE.

Cover of Value Book

Two of my books, Ten Things You Need to Know as In-House Counsel – Practical Advice and Successful Strategies and Ten (More) Things You Need to Know as In-House Counsel – Practical Advice and Successful Strategies Volume 2, are also on sale on the ABA website (including as e-books).

I have published two other books: The Evolution of Professional Football, and The Slow-Cooker Savant.  I am also available for speaking engagements, webinars/CLEs, coaching, training, and consulting.

Connect with me on Twitter @10ThingsLegal and on LinkedIn where I post articles and stories of interest to in-house counsel frequently.  

“Ten Things” is not legal advice nor legal opinion and represents my views only.  It is intended to provide practical tips and references to the busy in-house practitioner and other readers.

If you have questions or comments, or ideas for a post, please contact me at sterling.miller@sbcglobal.net, or if you would like a CLE for your in-house legal team on this or any topic in the blog, contact me at smiller@hilgersgraben.com.

[1] Which, as most long-time readers know is the origin story of the “Ten Things” blog format.

[2] I wonder what happened to Clippy.  Hopefully, a miserable and painful demise or on permanent exhibition at the Smithsonian of Shitty Ideas (right next to Kellogg’s Crunchy Loggs breakfast cereal and Theranos).

[3] Or work at Newsmax.

[4] For a list of thirty-three generative AI tools you can try for free, check out: Here Are 33 Generative AI Tools You Can Try for Free.

[5] See Jennifer Gibbs’ excellent article Lawyer Discernment is Critical in the World of AI.

[6] If I were a young in-house lawyer today, I would be learning about ChatGPT and positioning myself to be the department expert.  See As AI Advances is Prompt Engineering the Next Crucial Legal Skill?

[7] Sorry for all the adverbs.  Ernest Hemingway must be turning in his grave – assuming he’s a subscriber.

[8] Here are several articles that set out excellent ChatGPT prompts for lawyers: The Power of ChatGPT Prompts for Lawyers, The Best ChatGPT Prompts for Lawyers, Complete ChatGPT Guide for Lawyers: Top 20 Essential Prompts, and ChatGPT and Legal Writing: The Perfect Union?

[9] See also https://fellow.app/blog/productivity/using-chatgpt-for-meeting-minutes-a-better-alternative/.

[10] For ease, I am going to cite to the ABA Model Rules of Professional Conduct here, but you should always focus on the rules applicable to where you actually practice.

[11] Here is a great overview of the ethics issues and generative AI tools, The Ethical Implications of Using AI Tools for Texas Lawyers (covers Model Rules and Texas rules).  Being in Texas, I have this article bookmarked.

[12] See ChatGPT Has Severe Memory Limitations: Judges, Arbitrators and Commercial Litigation Lawyers, Your Jobs Are Safe for Now and The Memory Limitations of Large Language Models in Legal Analysis.

[13] See ChatGPT – Privacy Nightmare or Helpful Tool? or Data Protection in AI Chatting: Does ChatGPT Comply with GDPR Standards?

[14] See Harvard Business Review, The New Risks ChatGPT Poses to Cybersecurity.

[15] See, e.g., ChatGPT and the Future of AI Regulations.

[16] As Aldous Huxley wrote in Brave New World, “Stability isn’t nearly so spectacular as instability.”  Or as I say, “Instability = opportunity for in-house lawyers.”

[17] Check out Crafty Counsel’s series on how in-house lawyers are using ChatGPT.  Plus, it’s a great website overall for in-house lawyers. Worth the read!

[18] A lawyer recently used ChatGPT to draft a brief and filed it with the court.  Apparently ChatGPT made up cases and citations.  The lawyer didn’t review the output (which is crazy) and stated that he didn’t realize ChatGPT would make stuff up.  Even assuming the lawyer was that clueless, failing to check citations in a brief generated by ChatGPT is the definition of malpractice in my mind.  Don’t be that guy.  See, New York Times, Here’s What Happens When Your Lawyer Uses ChatGPT.

[19] Thanks to my friend Tony Bacon, GC at Hunter Fan, for this tip!

[20] Check out these excellent YouTube tutorials: ChatGPT for Law and an Introduction to Legal Prompt Engineering and ChatGPT for Lawyers: 101 Disruptive ChatGPT Prompts for Lawyers.  Indeed, YouTube is a great place to learn more about how to best use ChatGPT and other generative AI products.

[21] See New York Times, Microsoft Calls for A.I. Rules to Minimize the Technology’s Risk.