AI, ‘AI washing’ and the case for using your legal brain

Posted by Jim Hitch, chief executive of Legal Futures Associate Casedo

Hitch: Computers cannot yet replace the human brain

We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run. This has become known as Amara’s Law and the first part neatly sums up Rodney Brooks’ seven deadly sins of AI predictions.

The buzzword ‘artificial intelligence’ (AI) has remained steadfast within conversations around legal tech since I co-founded Casedo in 2017 and doesn’t appear to be going anywhere soon. It has promised us much and, whilst its benefits have been tangible, they are not intelligent services.

The fact is you cannot currently outsource intelligent legal work to something that’s not human. Of course, it depends on a definition of ‘intelligent’, but I would take it to mean not simply an ability to do a lot of calculations very quickly, the brain is much more than infinite monkeys randomly compiling the complete works of Shakespeare.

You can, however, outsource grunt work to computers. In law, contract grunt work used to be carried out by trainee solicitors, or banks of hired hands reading contracts, and this is still often the case today; outsourced by solicitors, barristers and sundry partners so they can get the data they need in order to come to conclusions/decisions that their clients are looking for. This is something that can readily be done by computers.

There are two things going on here, firstly whether AI itself actually exists and secondly, whether the businesses that purport to be selling AI services (whatever AI actually means) are actually ‘AI washing’.

Is AI a misnomer for the valuable services ‘AI’ legal technology businesses provide?

The debate as to whether AI is the correct name for these kinds of services has been going on for some time and will continue to rumble on for many years to come. Many of us use the term lazily, even though we suspect it might lack veracity. We’ve done the same at Casedo.

Technically, anything that can pass a modern equivalent of the Turing Test is ‘intelligent’. There have been many claims over the years, though the most recent famous example at Google was seen more as being good at deception rather than actual intelligence. Many in the science community are of the opinion ‘No, not even close’.

If a top-end Google computer is creating debate as to whether it is intelligent or not, commercial software, legal or otherwise, surely falls on the non-AI side of the fence.

However, this is not to say that the services many of these firms provide are not beneficial to the legal sector. They have streamlined processes and made some areas of legal services orders of 10 faster and cheaper for clients. Contract analysis software is a good example of this benefit whereas legal firms’ chatbots that claim to be AI driven can struggle with anything more that the most basic questions.

In his excellent summary, Alex Heshmaty gets to the point: “As it currently stands, any references to artificial intelligence within the context of legal technology should generally be viewed, in the words familiar to everyone who studied law, as ‘mere puff’.

“Practice managers considering investing in software should ignore ‘AI washing’ and focus on ascertaining the tangible benefits of any new technology.”

What is AI washing?

Now that we are all familiar with greenwashing and the like, this, I think, is self-explanatory. Brian Inkster sums it up rather well on The Time Blawg: “A marketing effort designed to imply that a company’s brands and products involve artificial intelligence technologies, even though the connection maybe tenuous or non-existent.”

To be clear, before sitting to write this article, I thought there was some debate as to whether AI is already available or not, but the more I have read, the clearer it seems that the consensus from those who are not interested in flogging us their wares is that no, AI is not here and won’t be for the foreseeable future.

So AI washing is in fact any talk about having AI technologies in a product.

At one end of the scale, this could be seen as forgivable (I’m not so sure); AI is the term used for a certain type of product, so if you are selling that type of product you need to use it to set yourself alongside your competitors.

What is less so is the deliberate ‘AI-ing’ of existing products in order to shift more units. This is AI as ‘magic’, as defined thus by Alex Hamilton’s glorious LawTech Glossary: “Artificial intelligence – A term for when a computer system does magic.

“‘General’ artificial intelligence refers to thinking computers, a concept that for the foreseeable future exists only in science fiction and lawtech talks.

“‘Narrow’ artificial intelligence refers to a limited capability (albeit one that may be very useful) such as classifying text or pictures, or expert systems. Discussions of AI that blur general and narrow AI are a good indication that you are dealing with bullshit.”

Narrow AI refers to single task systems such as chatbots, Siri or Alexa. Anyone who owns or has had to deal with one of these infuriating if useful devices knows that they are not intelligent.

Essentially, if you see AI in the marketing material for a product you are considering, it may be a great product, but it’s not AI, it’s AI washing, whether within the generally understood definition of AI or not. These systems may be great, but they are not intelligent.

The case for using your legal brain

Computers are great and many of these services that are labelled with the marketing hook of AI are also fantastically useful even if mis-labelled (unless simply understood as ‘magic’, as above).

But they only work when they work with the human brain; these services remove the drudgery leaving human to utilise, as Tim Harford puts it, “distinctly human stills”.

He wrote: “A task-based analysis of labour and automation suggests that jobs themselves aren’t going away any time soon – and that distinctively human skills will be at a premium. When humans and computers work together… the computers handle the ‘routine, codifiable tasks’ while amplifying the capabilities of the humans, such as ‘problem-solving skills, adaptability and creativity’…

“We’re human by virtue not just of our brains, but our sharp eyes and clever fingers.”

And this is where is comes back to. Since early humans started using tools, those tools have added to the human experience and to what can be achieved.

Until artificial systems can replicate our uniquely human skills, we must be careful not to rely on systems that claim intelligence because we may be relying on them for an intelligence they don’t have.

The brain remains the universe’s most complex structure. We should use tools to bring out the best in it, not to replace it.


Leave a Comment

By clicking Submit you consent to Legal Futures storing your personal data and confirm you have read our Privacy Policy and section 5 of our Terms & Conditions which deals with user-generated content. All comments will be moderated before posting.

Required fields are marked *
Email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Loading animation