paint-brush
ChatGPT Can Be Great or Horrible for Private International Lawby@asainidis
188 reads

ChatGPT Can Be Great or Horrible for Private International Law

by Alexandros SainidisMay 3rd, 2023
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

Many people are apprehensive about the impact of AI. While some have theorized that AI could eliminate the need for certain tasks or jobs, we are undeniably witnessing significant advancements in the availability of mainstream technology. What sets ChatGPT apart from other AI experiments thus far is its ability to handle language and conversations effectively. It is a powerful tool when it comes to processing languages. Casual users of ChatGPT often praise its capacity to engage in thoughtful and impressively intelligent human-like conversations, although mostly in an 'affirmative' tone. Admittedly, it does make mistakes or occasionally provides incorrect answers in a convincing manner. Nevertheless, as a scaled-down version of more advanced technology, it demonstrates just how far "language" has come in the realm of AI.
featured image - ChatGPT Can Be Great or Horrible for Private International Law
Alexandros Sainidis HackerNoon profile picture

The year 2019 ended with a disease outbreak that continues to cause difficulties and inconveniences, albeit to a lesser extent. In contrast to this economically destructive factor, 2022 concluded with a small Santa gift called "ChatGPT." This development piqued my interest, particularly in regard to its potential implications for globalization in the form of Private International Law.


Many people are apprehensive about the impact of AI. While some have theorized that AI could eliminate the need for certain tasks or jobs, we are undeniably witnessing significant advancements in the availability of mainstream technology. What sets ChatGPT apart from other AI experiments thus far is its ability to handle language and conversations effectively. It is a powerful tool when it comes to processing languages. Casual users of ChatGPT often praise its capacity to engage in thoughtful and impressively intelligent human-like conversations, although mostly in an 'affirmative' tone. Admittedly, it does make mistakes or occasionally provides incorrect answers in a convincing manner. Nevertheless, as a scaled-down version of more advanced technology, it demonstrates just how far "language" has come in the realm of AI.


Languages, in the broad sense, are not limited to words. You can express different parameters of a vase better in different languages in the following ways:


1. Using literal, objective adjectives (white, tall, colourful)

2. Metaphorical phrases to express the feelings we associate with it (a menacing vase, which makes you feel it contains an ancient curse which will possess an omniscient AI and rule the world once broken)

3. Math and geometry (this vase is 15 centimeters tall and has a radius of 5 centimeters)

4. Draw it on a piece of paper – this is why sketching could be regarded as a form of communication

5. Other ways I am possibly not aware of, at the moment


So is Law a Language?


Yes, closer to a programming one. It attempts to program human behavior by using incentives (rights, rewards) and counter incentives (prohibition, punishment). It sets conditions and most importantly a hierarchy of conditions. Private International Law asks the following question: if every country has its own set of rules, which ones do we follow when there are many countries involved because some individuals got more international than they could chew?


And what can a ChatGPT do for that?


Imagine instead of checking the laws and cases of each country, as well as the effects of bilateral and multilateral agreements you simply connected a ChatGPT or a ChatGPT copy/improvement to a large and up-to-date legal database. This ensures that the engine has access only to verified information and eliminates noise created by irrelevant or unreliable sources. With one input of the events that took place and data from documents, the LawGPT can easily compare and see what applies, within seconds. This means that legal consulting can change overnight. It would not come as a surprise if the leading global law firms have created a similar internal database already. Lawyers are notorious for storing their own rabbit-holes of experience and the AI-assisted use of legal data could easily be the outcome of a digital transformation policy implementation. So how should a lawyer differentiate themselves in terms of quality of their services, if LawGPT can be very effective?


In the end, it’s about Soft Skills. Unsurprisingly, clients prefer to be around humans. Power of speech is not only about proving a point at court. It’s about handling your customers by genuinely being interested in what they are dealing with. This can be demanding and draining, which is why many lawyers avoid it as much as it can – but that’s how you score points. The number one problem is lack of proactivity. Lawyers lose too much energy appeasing an agitated client instead of being proactive. For example, if the client needs to wait 20 days for a divorce paper coming from Sri Lanka and they don’t know about it yet, would they be more angry:


a. When the lawyer calls the client and communicates it since day 1 or

b. When the client calls the lawyer on day 10


If the client calls first, chances are that the client will think that they totally forgot about it and only filed an application now that they were reminded. In a chess match with a client, always choose white.


Others forget culture. Is it okay to call a client of an Arab state at any point of the day? Do we expect all public administrations around the world to follow or accept the same patterns of communication? What about foreign lawyers; are there any customs or protocols to follow when going against Malagasy ones? You can answer those far better than a robottorney, because it’s not just about words. It’s about behavior.


Also published here.