This is a ChatGPT Free Law Blog.
It is a real Law Blog with no AI used to create the content.
No Hallucinations in a ChatGPT Free Law Blog
The content is created by real human beings using critical thought processes with no hallucinations involved.
As was the case in the heyday of blawgging.
We appear to have entered another 5 year cycle of over hype about the use of AI in Legal Tech.
No AI in a ChatGPT Free Law Blog
It is, I believe, important for realistic balanced views of the tech being employed to be presented. And it needs to be clear to the reader that AI didn’t write the content.
Some platforms for writing/promoting law blogs are sadly incorporating and encouraging the use of ChatGPT. They have been caught up in the hype. They appear to think that all Legal Tech, including their own, needs to incorporate this latest fad. Just like the Legal Chatbots who were once ‘trained’ by IBM’s Watson!
You can rest assured that at The Time Blawg we will be taking no part in that. We are and will always be a ChatGPT Free Law Blog. But that is probably obvious as this post has clearly not been written in the style of a Shakespearean sonnet!
Chat GPT and Lawyers
I will, when I get the chance, do a full and detailed post on ChatGPT and lawyers. That may contain some examples of ChatGPT at work but those will be put in quotes and clearly labelled as the work of ChatGPT. That is how it should be. I hope other real law bloggers follow suit and declare their Law Blogs to be ChatGPT free.
Chat GPT Rabbit Hole
If you are a law blogger please don’t get lead down the ChatGPT rabbit hole. Write your own authentic content and avoid hallucinating.
Image credit: Harvey (1950 film) © Universal Pictures
Reactions on Social Media to ChatGPT Free Law Blog
On LinkedIn the following comments have been made:-
Graeme Johnston (Software to map work – before that a lawyer):
I’m quite tempted to propose a set of laws about hype in this field, inspired by certain physical laws. For example, 1st law: the total hype in the field can never decrease but may change in form. More AI hype may involve less blockchain, for a while anyway. 2nd law: every new hype has an opposite but not-quite-equal anti-hype. 3rd law (consequential): the total hype in the field increases over the long run.
Me:
You should – or have you just done so! Blockchain seems to have almost been forgotten about and perhaps replaced already by this latest fad?
Sarah Glassmeyer (LexPunk):
Brian Inkster – After listening to yet another discussion on the future of generative text in law on a podcast yesterday, I actually thought to mysefl “I think I miss the blockchain discourse.”
Graeme Johnston:
Brian Inkster – Its promoters are currently hunkered down figuring exactly how to play the SVB collapse to reboot the hype. I expect next week there will be quite a bit!
Alex Smith (Global Search & AI Product Lead (Senior Director) at iManage):
It creates bullshit but you shouldn’t bullshit about what it does.
https://www.ftc.gov/business-guidance/blog/2023/02/keep-your-ai-claims-check
Steve Brett (E3CT – Compliance training can be engaging and effective):
Alex Smith if they say AI it’s written in PowerPoint. If they say machine learning it’s written in Python. Not sure who said this but it is so true.
I’ve seen a number of products “driven by AI” that at best are a couple of layers of nested if then statements….
Alex Smith:
From the last wave of AI … issue is even bigger bullshitters involved rolling of a bloody nose on web3. https://archive.md/bMaBW#selection-177.0-177.97
“The core feature of a B.S.-industrial complex is that every member of the ecosystem knows about the charade, but is incentivized to keep shoveling. It’s not so much that we reach a point where we convince ourselves our bullshit is true; it’s that the difference between truth and bullshit has become purely semantic. The definition of something, like artificial intelligence, becomes so jumbled that any application of the term becomes defensible.”
-+-+-+-+-+
Victoria Moffatt ChartPR (Non-practising solicitor, founder and MD of LexRex Communications. Legal PR, law firm marketing, Chambers and Partners and Legal 500 specialist.):
Hard agree. I do think there is a place for this kind of tech, but if it works as I *think* it does, then utilising it will enable you to create content that probably takes ideas and concepts from existing content. Meaning your content will become a mixture of content that’s already out there.
Not ideal if you are trying to establish a brand and / or stand out from the crowd. And I’ve never yet met a lawyer that isn’t trying to do one or both of these things.
Me:
Indeed Victoria Moffatt ChartPR. Those using it for content generation will end up with bland unauthentic content that will be more likely to affect their brand negatively than anything other.
You may recall from a decade or more ago the blawgers v flawgers debate: https://thetimeblawg.com/2012/01/04/flawging-lawyers-come-out-of-their-chambers/
Whilst I take the view that all blogging has a flogging aspect, it will be clear, I think, that those lawyers using ChatGPT to create their law blogs will be this decade’s outright flawgers.
But when it comes to marketing your brand and selling your services I believe those with valuable content (cc Sonja Nisson and Sharon Tanton) that is authentic and original will always trump the ChatGPT flawgers.
Victoria Moffatt ChartPR:
Gosh I’d forgotten that this was ever an argument haha! I definitely think a mix of blogging and ‘flawging’ is probably sensible (not sure what my equivalent is) – again ensuring that any tactics support the broader business strategy.
I need to go and have a play with ChatGTP to be perfectly honest but I have been trying to keep up with the various comment pieces covering it.
Again, if you need quick content and it doesn’t matter / you don’t care (not always the same ) / you don’t have time / resource etc to draft original content then I can see why it would be tempting to use this tech. As ever, shades of grey and realities of resource management + pragmatism.
Me:
Indeed Victoria. And clearly anyone using it without double (or maybe triple) checking the accuracy of its output does not care.
-+-+-+-+-+
Steve Brett (E3CT – Compliance training can be engaging and effective):
I’m struck by Arthur C Clarke – any technology sufficiently advanced will appear like magic.
AI / machine learning are very powerful tools but as ever a hammer is little use if you need to fix a screw! It will work(ish) but could end very badly.
I see a big issue is that many people don’t understand how this technology works and more importantly the limitations. A key problem being the training data used.
Chat GPT has been trained on a massive data set but it’s biased. If you have a problem within the constraints and you understand what the algorithms are optimising for, great.
If not you will either get rubbish out or worse something you thing is golden but in reality is just a glitter encrusted turd.
Me:
Indeed Steve. Imagining magic is the second deadly sin of AI predictions as set out by Rodney Brooks and covered by me in relation to LegalTech here: https://thetimeblawg.com/2018/10/01/travels-through-the-blawgosphere-2-artificial-intelligence-and-law-robots-replacing-lawyers/
The legal futurists’ excitement over ChatGPT probably commits all seven of the deadly sins. I will look at that in more detail when I do a full blog post on ChatGPT and lawyers 😉
-+-+-+-+-+
Philippos Aristotelous (Business Strategist I Trainer I Consultant):
Insightful comment and position. Here are some additional thoughts:
There is no one-size-fits-all argument concerning ChatGPT as it is not one thing for everyone.
Embracing new technologies like AI in law firm blog writing can increase productivity and efficiency, improve accuracy, and demonstrate a commitment to innovation.
History has shown that disruptions ultimately lead to progress and cultural evolution, as seen in the rise of the printing press and digital music.
AI tools like ChatGPT can similarly unlock new possibilities for productivity, accuracy, and innovation in blog writing. From merely proofreading to generating content from scratch, there is a moral spectrum of use, and the dilemma of whether to use ChatGPT or not is not a simple one.
It’s similar to the dilemma of using a compass versus GPS: both tools can help navigate, but depending on the situation, one may be more appropriate than the other.
By approaching AI responsibly and ethically, law firms can navigate this dilemma and improve their productivity, accuracy, and creativity in blog writing while demonstrating a commitment to innovation in a rapidly changing industry.
Me:
Thanks Philippos Aristotelous. The “curious” emoji reaction 🤔 appears to have been removed by LinkedIn as a ‘like’ option otherwise I would have used that 😉
Thoughts on your points:-
“Can increase productivity and efficiency”: Yes, but at what expense? As Victoria Moffatt ChartPR pointed out, your content will end up a mixture of content that is already out there. You will not stand out.
“Improve accuracy”: I doubt that, given where ChatGPT technology stands at the moment. It is well known that it hallucinates. In other words it just makes things up. The time spent in checking and correcting its accuracy might be better spent researching source material.
“Commitment to innovation”: You can show that commitment without necessarily picking up and playing with every shiny new toy. There is a lot to be said for lawyers concentrating their efforts on existing proven technologies e.g. document automation perhaps? #bringbackboring
However, I am not ignoring the fact that ChatGPT is probably a step in the direction of AI becoming, in some respects, more useful to mankind. But I think we are a long way away from seeing it being the panacea that many are making it out to be today.
Philippos Aristotelous:
Indeed the future iterations will for sure answer a lot of these questions.
Me:
And as long as we don’t kid ourselves that this will magically happen with ChatGPT4 (which if anything may make things worse with its language conversions and video capabilities). To do so would be to commit the 1st and 7th deadly sins of AI predictions and maybe one or two in-between 🙂
-+-+-+-+-+
Heather Suttie (Legal Market Strategy and Management Consulting — Global to Solo I BigLaw to NewLaw)
Authenticity is the basis of every strong brand and unique personality. This is why ChatGPT or any AI-inspired communication will never replace an individuals voice and way with words. Alternatively, should you wish to be replaced by a robot, here’s your chance. This will never happen to us, Brian.
Me:
Thanks Heather Suttie. I very much agree with you.
-+-+-+-+-+
John McCarthy (Law Firm Profit Sherpa):
Interesting post Brian Inkster, I think AI has/ will have its place.
Like all useful technology, it’s about knowing how and when to use it.
Tech & AI are an integral part of our lives now, so it’s about controlling & using it to the best effect for us as individuals and our businesses.
Me:
Indeed John.
I think those saying lawyers must use ChatGPT now or they will be left behind are misleading those in the profession who might listen to them.
The same was said about blockchain, cryptocurrencies, Clubhouse and the metaverse! It was also said about AI at least 5 years ago and probably 5 years before that.
In some cases it is better to be a laggard than an early adopter 😉
-+-+-+-+-+
Stephen Gold (Principal at Stephen Gold Consulting: Helping Lawyers Build Great Firms):
Thoughtful as always, Brian.
We need to be careful though that we don’t overstate the sacredness of originality. There are very few thoughts that are genuinely new under the sun and haven’t been influenced by others.
It’s objectionable to download ChatGPT content and unthinkingly regurgitate it. But to use it as a source of ideas, or fact checking, on which one can build one’s own ideas is surely no more objectionable than using books or quality journalism, subject to the usual rules of attribution.
Victoria Moffatt ChartPR (Non-practising solicitor, founder and MD of LexRex Communications. Legal PR, law firm marketing, Chambers and Partners and Legal 500 specialist.):
This is where I can particularly see its value. Research and idea creation.
Me:
Thanks Stephen.
Agreed that it might have a use as a source of ideas.
At the moment using it for fact checking may be problematic due to it hallucinating!
Unlike a Google search it doesn’t tell you its sources of reference.
So if you ask it who Brian Inkster is it might well tell you that he is a lawyer in Scotland. But it might also tell you that he is married to the American professional golfer Juli Inkster. It might go on to tell you that he played hockey in Canada.
Each of these facts is true. But each is only true of one particular Brian Inkster. ChatGPT is finding information in its database about three different people all named Brian Inkster: one Scottish, one American and one Canadian. It can’t tell the difference between all three and so just assumes they are one and the same and mashes all of the information together.
What we really need to see disclosed is the sources of the information it is relying on so that we can delve deeper into the source material and check the accuracy of its output.
But if we are having to do that on a regular basis are we not better just doing a Google search and building our own content from the information we read and decide (based on the source) that we trust?
Stephen Gold:
Brian, you’re right about the fact-checking shortcomings. It gave me the right answer to “What is the capital of Scotland?”, but in answer to, “Who is the British Prime Minister?” It said:
“As an AI language model, I don’t have real-time information, but as of my knowledge cutoff date of September 2021, the British Prime Minister was Boris Johnson. However, it’s possible that there has been a change in leadership since then.”
Well indeed….
Me:
Indeed! At least two!
-+-+-+-+-+
Antti Innanen (⚫ Ⓜ️ 🍄):
Prompt: Write a pessimistic LinkedIn post about how your law blog will never use ChatGPT-created content. Include ideas ”content is created by real human beings”, ”there are no hallucinations involved” and ”human-generated content is more authentic”.
You are not fooling me 🤪
Me:
Clearly that will work well once my blog post has been added to the ChatGPT data set and someone does that prompt and plagiarises my original and authentic post! 😉