Are robots going to steal our job ?

This is a title

What is artificial intelligence?

According to the definition of Wikipedia, artificial intelligence (AI) is a scientific discipline seeking methods for solving problems of high logical or algorithmic complexity. In the common language, it refers to devices imitating or replacing humans in certain implementations of their cognitive functions.

The starting point of the concept of artificial intelligence dates back to the 1950s, but it was mainly mediated during recent events. In 2016, when Google DeepMind’s Alphago program beat Go World champion Lee Sedol. Already in 1997, Deep Blue, IBM’s supercomputer had defeated world chess champion Garry Kasparov.

The recent dynamic of AI is enabled by the development of new information processing technologies: the computing power that continues to follow Moore’s law, as well as data storage capabilities and algorithmic techniques such as Deep Learning, which now rely on neural networks, enabling machines to assimilate repetitive processes without the need for programming, but only by reproducing the execution of tasks done by a human.

It is an indisputable fact: artificial intelligence will generate transformations in all sectors of human activity. Some observers predict that artificial intelligence will even be the number one competitiveness factor in the world of tomorrow, whatever the theme.

Have you already asked yourself: “Are robots going to steal my job?”

Several studies show that by 2040, 40 to 50% of professions will have been affected by technological developments. Already today, some robots and dedicated software already have a profession in its own right. What about our (your) area of expertise?

The list of occupations is long: secretaries, accountants, journalists, soon doctors and lawyers: these so-called white-collar jobs could soon be done by machines. At Oxford University, researchers Carl Frey and Michael Osborne, announce that in the US, one job in two could be automated by 2035.

We have become accustomed to seeing the disappearance of manual, highly repetitive and low- skilled jobs, such as the work of factory operators, freight agents or assemblers. Today, even moderately high qualified professions are also threatened: insurers, analysts, accountants … even sports referees.

We are beginning to see algorithms appearing on boards of directors to shed light on the opinions of members and thus inform investment decisions. Confidence in the algorithm is sometimes such that if the algorithm refuses an investment, the board will follow its advice and will not take the risk.

(See the example of Vital used by Deep Knowledge Ventures in Hong Kong).

Some software, like Quill, has the ability to write like journalists. To write simple and factual content, no need to pay a journalist. The Associated Press already uses this system on a large scale (1,000 articles per month to list the financial results of US companies). In Europe too, we see flourishing software solutions of this type, adapted to the decryption of international languages often a little more complicated than business English …

The trend is also visible in France, where Roland Berger estimates that 42% of jobs will be automated within twenty years. Three million jobs could be destroyed in ten years, and the French unemployment rate would reach 18%, including jobs with high intellectual content.

One of the most widely known AI systems is called Watson. This supercomputer, developed by IBM, is, among other things, capable of performing rapid and reliable medical diagnostics.

Beyond Moore’s law, what determines the success of this revolution is called ‘Big Data’. Companies, public actors and soon all the inhabitants of the planet will have to be absolutely transparent so that as much useful data as possible can be available, exploitable (and beneficial to everyone)!

The development of digital has meant deep changes for companies, for example in the insurance market, because traditional insurers can no longer rely solely on their limited internal resources, who do not always have the specific skills to operate and perform mass processing and address complex data analysis challenges.

This paradigm shift has been integrated since the beginning of the decade into the relationship between insurance actors and the general public.

The development of digital technology, however, has not yet been fully adopted in professional procurement services, even by technology purchasing specialists. If it is an area that can benefit from immeasurable benefits from automation, and without taking on the employment of human operators, it is the management of the life cycle of contracts (so called CLM)!

A contract is formalized by a legal document that creates obligations between two parties. Everyone is always on contract every day and at every level, for example when buying bread from the baker or subscribing to a new ultra-competitive service.

Consumer contracts do not really interest us at this stage: let’s look at the service contracts that make companies dependent on hi-tech providers (telecom operators, IT service companies, AI developers, etc.) and also Your service contracts to Your customers.

A contract in the first place is what is visible, i.e. everything that is written in small or bold print on a document that you will sign, but it is also and especially what is not seen, which depends on local, US, European, Chinese (?) laws, which already apply in a large number of cases during trade and which are often more powerful than what you can put in writing.

It is necessary to distinguish what are called public order requirements (one can write what anyone wants in a contract, but if it is contrary to the tax, customs, consumption laws, the contract will be worthless) and then the suppletive laws coming from the Civil Code that apply in everyday life and never ask to sign a paper, such as when you buy your newspaper will apply.

 

Before looking at contracts, you have to understand that there are a multitude of laws and jurisprudence that can be applied.

At least four major sections must be present in a contract:

– The object: the object specifies the requirements of the customer. It includes the commercial offer or the proposal received by the customer from the vendor.

– Duration and method of renewal: most contracts are offered for a relatively short period (1 to 3 years), but often with a tacit renewal clause for the same term, with no possibility of termination before the term. As a result, many companies find themselves paying for services that they have not used for a long time.

– The territory: the territory to which the contract applies and therefore the laws on which it depends must be clearly stipulated. Signing a restrictive contract in a country where the law is permissive will not protect you.

– Price and delivery date: Notions that seem obvious, but often neglected after having transpired on more exciting legal concepts. If you mention a delivery date, your supplier must respect it and what will happen if it is not to be provided! If you have approved a tariff, it must be physically applicable.

You (or your purchasing department) have taken care of negotiating an extremely sophisticated service contract with a major service provider (IT, telecom or other). For the buyer, it is a great achievement. But in reality, this is only the beginning of problems: a contract does not come to life until the moment of its signature.

In the majority of companies, everything stops at contract signature, and control is then given to the operational level, who will exploit the service without ever worrying about the relevance of billing, or even often about compliance with contractual commitments.

Worse: as mentioned above, the service will someday be abandoned by the operational staff, but the contract signed with tacit renewal will continue to be invoiced until an accountant raises the question of “what’s the point?”

Now you understand: the life of a contract starts at its signature, and stops at its denunciation. Between the two, for years, it will be necessary to follow it, to check that the services rendered are legal, conform to the specifications, priced at the agreed price, and it will be necessary to anticipate the deadline for either renegotiating or stopping an obsolete service.

Do your buyers do this work? Congratulations you are lucky! Keep them!

Nobody does this job in your enterprise! Why? You’re just in the average. Because the complexity / value added ratio is perceived as a waste by negotiating professionals who are more passionate about writing new contracts with new suppliers.
The lawyers? Too fascinated by the resolution of complicated cases, they have no interest in assuming operational control of signed contracts.

 

That’s why contract lifecycle management is an area that can benefit from robotization without robbing the work of any human operator! It is no one’s job!

It is only a question of entrusting a computer an automation mission that no human wants to execute because they are too tedious and where the risk of human error is very important. Who would like to expose themselves to the thunderbolts of management just because we did not know that a legal error would lead to an operational catastrophe?

Today, most legal departments have acquired “Contract Library” software. This is remarkable because it gives legal directors the feeling that the problem of managing the life of contracts is under control.

Let’s face it, contract library management software is inoperative. They only serve to give legal directors a good conscience. Just can they guarantee a recording of the documents in their final form, and it is not even certain. The boring task of recording contracts in the database is generally given to an intern who barely has a clue of what s/he is doing. Often times, only a scanned image of the signed document is kept, thus losing any ability to perform searches on key words.

The solution is to let robots perform an automated analysis of the document before signature. Such robots would use artificial intelligence to understand the meaning of complex legal language and transform it into reference formulas and propose possible renegotiations for achieving a form of “perfection”, while respecting good practices.

Certain elements of the contracts are simple to detect, such as term dates, renegotiation deadlines, price revision dates and associated formulas, penalty calculation formulas for non- compliance with commitments, etc. All this is often too complicated for a trainee, and a contract library left with a lot of poor quality data is worthless.

The set of mathematical object models integrated into the contracts in the form of service levels, KPI, credits / earnings, rates, etc. are often represented in contracts as simple arithmetic formulas. It is possible to extract these objects from the contract and convert them into computer algorithms so that when the performance review takes place, the data produced can automatically challenge the performance of the supplier.

New generations of contract management technology take a very different approach to traditional contract life cycle management. By using advanced technological concepts such as natural language processing, artificial intelligence, machine learning, workflows, intelligent algorithms, etc., these contract management systems do more than create and store contracts, enabling the achievement of the expected results.

Adding to the Contract Life cycle Management (CLM) advanced contract management capabilities increases the ROI of its contracts.

Moreover, the generation of warnings that would allow a possible recourse is often overlooked, and many companies find themselves hired for hundreds of thousands of Euro in contracts that no longer serve anything, just because they have been neglected.

Beyond automated reading and recording, one can imagine a robot that offers proposals for improvement of contracts, before they are signed:

– To offer formulations based on best practices, based on the reading of thousands of examples – Detecting legal errors and suggesting corrections
– Quantifying legal risks and proposing alternatives to reduce it, in particular by eliminating non- standard clauses.

Author: Gérard Thierry

Gérard is a frequent collaborator, and a senior advisor to UNEGO. For more information, please contact UNEGO at info@u-nego.com 

Posted on 13/08/2017 in Artificial Inteligence, CLM

Share the Story

Back to Top