Natural language processing: Difference between revisions

From The Jolly Contrarian
Jump to navigation Jump to search
No edit summary
Tags: Mobile edit Mobile web edit
No edit summary
Line 1: Line 1:
{{a|tech|}}One of the [[Holy Grails of reg tech]] is [[natural language processing]]: some varieties of the same thing: a machine that reads contracts for you. This could come in the following articulations:
{{a|tech|}}One of the [[Holy Grails of reg tech]] is [[natural language processing]], a handful of varieties of the same thing: a machine that reads {{t|contract}}s for you.  
 
===Examples===
*'''Data extraction''': Crawling over your portfolio of 40,000 {{isdama}}s<ref>You know, the ones printed on faded waxy fax paper and languishing in filing cabinets around the trading floor; the ones scanned into a 57 MB tiff file along with three amendments, forty pages of specimen signatures, a power of attorney, hand-annotated emails from Credit and the five key pages of the Schedule missing; the ones that are misfiled as Swiss [[rahmenvertrag]]s; the ones that are just not there '' at all''.</ref> to extract the 60 key [[trading]] and [[credit]] terms out of them that the firm neglected to collect over the last 25 years while it was signing them up;
*'''Data extraction''': Crawling over your portfolio of 40,000 {{isdama}}s<ref>You know, the ones printed on faded waxy fax paper and languishing in filing cabinets around the trading floor; the ones scanned into a 57 MB tiff file along with three amendments, forty pages of specimen signatures, a power of attorney, hand-annotated emails from Credit and the five key pages of the Schedule missing; the ones that are misfiled as Swiss [[rahmenvertrag]]s; the ones that are just not there '' at all''.</ref> to extract the 60 key [[trading]] and [[credit]] terms out of them that the firm neglected to collect over the last 25 years while it was signing them up;
*'''Legal agreement review''': algorithmically scanning standard-form {{t|contract}}s<ref>To date, only one any one has successfully managed is the one that no-one really cares about: the [[confidentiality agreement]]</ref> to identify key terms and risk provisions and save human lawyers from that tedious chore;
*'''Legal agreement review''': algorithmically scanning standard-form {{t|contract}}s<ref>To date, only one any one has successfully managed is the one that no-one really cares about: the [[confidentiality agreement]]</ref> to identify key terms and risk provisions and save human lawyers from that tedious chore;
*'''[[Chat bots]]''': An online, chat buddy to whom [[Sales]] can basic legal questions, thereby saving [[Sales]] the aggravation of having to talk to the [[legal eagles]], and [[legal]] the utter [[tedium]] of having to answer the exact same question to the exact [[Sales]] person three or four times daily.
*'''[[Chat bots]]''': An online, chat buddy to whom [[Sales]] can basic legal questions, thereby saving [[Sales]] the aggravation of having to talk to the [[legal eagles]], and [[legal]] the utter [[tedium]] of having to answer the exact same question to the exact [[Sales]] person three or four times daily.


Now reading any text involves judgment, interpretation and negotiation of ambiguity — and bringing to the text the reader’s own understanding of the legal background — while legal language is crafted to avoid ambiguity — {{maxim|there are no metaphors in a trust deed}} — there are still infinite ways of expressing the same idea, and if there is one part of the imagination a lawyer loves to stretch, it is inventing burlesque ways of saying simple things. Understand a well-formed English sentence is not just a matter of applying basic rules of language. It is a dynamic process.
Now reading any text involves judgment, interpretation and negotiation of ambiguity — and bringing to the text the reader’s own understanding of the legal background — while legal language is crafted to avoid ambiguity — {{maxim|there are no metaphors in a trust deed}} — there are still infinite ways of expressing the same idea, and if there is one part of the imagination a lawyer loves to stretch, it is inventing burlesque ways of saying simple things. Understand a well-formed English sentence is not just a matter of applying basic rules of language. It is a dynamic process. So expect [[natural language processing]] to be easier said than done.
 
And so it proves.
 
===Legal agreement review===
There is a well-known and widely feted [[natural language processing]] application<ref>Which shall remain nameless, though you don’t have to be a total ''nerd'' to know who we have in mind.</ref> which purports to save resources and reduce risk by performing a preliminary review of, say, [[confidentiality agreement]]s against a preconfigured [[playbook]].
 
The idea is [[triage]]. The application scans the agreement and, using its [[natural language processing]], will pick up the policy points, compare them with the playbook and highlight them so the poor benighted lawyer can quickly deal with the points and respond to the negotiation. The software [[vendor]] proudly points to a comparison of their software against human equivalents in picking up policy points in a sample of agreements. The software got 94% of the points. The meatware only got 50%. But in itself this may 
 
But it get the [[triage]] backwards. Rather than having the lawyer pick up the major points (the high value work) and then employing the [[AI]] to process and finalize the detail, it is the [[AI]] which picks up the major points and tasks the lawyer with completing the clerical work. For the process to be productive the lawyer must rely on the AI to have identified '''all''' salient points. Otherwise, the lawyer must read the agreement in full as a sense check. In practice, natural language processing is not sophisticated enough to allow this level of comfort, nonetheless lawyers are encouraged to trust it. Hence a buried risk.
 
Furthermore the reality is that many of the policy points in the [[playbook]] will be non-essential "perfect world" recommendations ("[[nice to have]] s") which an experienced negotiator will quickly be able to wave through in most circumstances.
 
But this software is designed to facilitate "right-sourcing" the negotiation to cheaper (ergo less experienced) negotiators who will rely on the playbook as guidance, will not have the experience to make a commercial judgement unaided and will therefore be obliged either to [[escalate]], or to engage on a slew of [[nice to have]] but bottom-line unnecessary negotiation points with the counterparty. Neither are good outcomes. Again, an example of [[reg tech]] creating [[waste]] in a process where investment in experienced human personnel would avoid it.
 
The basic insight here is that if a process is sufficiently low in value that experienced personnel are not justified, it should be fully automated rather than partially automated and populated by inexperienced personnel
 


{{ref}}
{{ref}}

Revision as of 15:49, 28 August 2019

The JC pontificates about technology
An occasional series.


Comments? Questions? Suggestions? Requests? Insults? We’d love to 📧 hear from you.
Sign up for our newsletter.

One of the Holy Grails of reg tech is natural language processing, a handful of varieties of the same thing: a machine that reads contracts for you.

Examples

  • Data extraction: Crawling over your portfolio of 40,000 ISDA Master Agreements[1] to extract the 60 key trading and credit terms out of them that the firm neglected to collect over the last 25 years while it was signing them up;
  • Legal agreement review: algorithmically scanning standard-form contracts[2] to identify key terms and risk provisions and save human lawyers from that tedious chore;
  • Chat bots: An online, chat buddy to whom Sales can basic legal questions, thereby saving Sales the aggravation of having to talk to the legal eagles, and legal the utter tedium of having to answer the exact same question to the exact Sales person three or four times daily.

Now reading any text involves judgment, interpretation and negotiation of ambiguity — and bringing to the text the reader’s own understanding of the legal background — while legal language is crafted to avoid ambiguity — there are no metaphors in a trust deed — there are still infinite ways of expressing the same idea, and if there is one part of the imagination a lawyer loves to stretch, it is inventing burlesque ways of saying simple things. Understand a well-formed English sentence is not just a matter of applying basic rules of language. It is a dynamic process. So expect natural language processing to be easier said than done.

And so it proves.

Legal agreement review

There is a well-known and widely feted natural language processing application[3] which purports to save resources and reduce risk by performing a preliminary review of, say, confidentiality agreements against a preconfigured playbook.

The idea is triage. The application scans the agreement and, using its natural language processing, will pick up the policy points, compare them with the playbook and highlight them so the poor benighted lawyer can quickly deal with the points and respond to the negotiation. The software vendor proudly points to a comparison of their software against human equivalents in picking up policy points in a sample of agreements. The software got 94% of the points. The meatware only got 50%. But in itself this may

But it get the triage backwards. Rather than having the lawyer pick up the major points (the high value work) and then employing the AI to process and finalize the detail, it is the AI which picks up the major points and tasks the lawyer with completing the clerical work. For the process to be productive the lawyer must rely on the AI to have identified all salient points. Otherwise, the lawyer must read the agreement in full as a sense check. In practice, natural language processing is not sophisticated enough to allow this level of comfort, nonetheless lawyers are encouraged to trust it. Hence a buried risk.

Furthermore the reality is that many of the policy points in the playbook will be non-essential "perfect world" recommendations ("nice to have s") which an experienced negotiator will quickly be able to wave through in most circumstances.

But this software is designed to facilitate "right-sourcing" the negotiation to cheaper (ergo less experienced) negotiators who will rely on the playbook as guidance, will not have the experience to make a commercial judgement unaided and will therefore be obliged either to escalate, or to engage on a slew of nice to have but bottom-line unnecessary negotiation points with the counterparty. Neither are good outcomes. Again, an example of reg tech creating waste in a process where investment in experienced human personnel would avoid it.

The basic insight here is that if a process is sufficiently low in value that experienced personnel are not justified, it should be fully automated rather than partially automated and populated by inexperienced personnel


References

  1. ↑ You know, the ones printed on faded waxy fax paper and languishing in filing cabinets around the trading floor; the ones scanned into a 57 MB tiff file along with three amendments, forty pages of specimen signatures, a power of attorney, hand-annotated emails from Credit and the five key pages of the Schedule missing; the ones that are misfiled as Swiss rahmenvertrags; the ones that are just not there at all.
  2. ↑ To date, only one any one has successfully managed is the one that no-one really cares about: the confidentiality agreement
  3. ↑ Which shall remain nameless, though you don’t have to be a total nerd to know who we have in mind.
Retrieved from "https://jollycontrarian.com/index.php?title=Natural_language_processing&oldid=32599"