Showing posts with label AI. Show all posts
Showing posts with label AI. Show all posts

Tuesday, February 9, 2016

Robots will force experts to find other routes to the top

FT | Andrew Hill | February 8, 2016





There is a point in any talk about the automated future of the professions when the audience visibly relaxes.


It comes when futurists concede that a few expert lawyers, consultants or accountants will still be needed, even after cheaper, more efficient computer systems have taken over many of their juniors’ tasks.


It happened last week at a lecture by Richard and Daniel Susskind, which the organisers claimed was the largest ever gathering of senior managers in UK professional services firms.

The father-and-son authors of The Future of the Professions predicted radical change in the sector. But the tense scepticism in the room dissipated as each senior partner or director quietly acknowledged he or she would be a survivor, even if algorithms and artificial intelligence swept away the consultant or solicitor in the next seat.

This cohort may well reach retirement unscathed — and without much incentive to alter how they work. As Richard Susskind told me afterwards, “it’s hard to convince a room full of millionaires that they have got their model wrong”. But change is coming. The main difference of opinion is over its pace and extent.

You can already ask Kim, a legal “virtual assistant” launched by Riverview Law, for help managing your caseload, or get Ross, IBM Watson’s “super intelligent attorney”, to research the entire body of law in seconds. But a crystal-ball-gazing report by The Law Society, the trade body for solicitors in England and Wales, expects the impact of this type of automation to level off by 2020.

The society’s Stephen Denyer told last week’s gathering that clients were not only looking for practical counsel, but for “negotiating skills, judgment, ethical standards, and reassurance about the direction they’re taking”.

Fine. But how will the senior partners of the future achieve that level of wisdom when machines are doing the tasks that allow them to build and hone their expertise?

Take financial journalism. I spent three years as a trainee building confidence and skill by churning out news about corporate earnings. This is precisely the type of report that, quite rightly, Associated Press now produces automatically, in partnership with a company called — ominously for all columnists — Automated Insights.

Another parallel is aviation, where crashes often trigger fears that autopilots are undermining human skills. Interviewed last year about the 2009 Air France crash, Delmar Fadden, Boeing’s former chief of cockpit technology, told Vanity Fair that, having automated 98 per cent of pilots’ routine work, “we really worry about the tasks we ask them to do just occasionally”.

The answer is not to halt the march of the robots. Indeed, technology is part of the solution. Novice astronauts are not trained by sending them on repeated costly moonshots. They practise the tasks and challenges they will face in carefully designed simulations, until they are finally ready for the launch pad.

As Prof Susskind points out, law students at the University of Strathclyde play out real-world legal problems in a fictional virtual community called “Ardcalloch”.

In the real world, professionals must recognise much of the work they hand to juniors is repetitive servitude, often imposed on the tacit assumption that, if they had to do it, so should the new generation. Clients may still prefer dealing with human experts but they do not much like paying for their juniors’ billable training.

Knowledge can be imparted in other ways, including simply by working closely, apprentice-style, with senior colleagues. I still value the guidance I received as a beginner from experienced editors and writers but I am not sure I needed to write five similar corporate earnings stories a day to achieve mastery. Newbies can acquire specific skills through working, under close supervision, on a sample of the basic tasks they once spent years slogging through.

Meanwhile, new roles will evolve. The Susskinds suggest that one could be the “empathiser”. A sympathetic human may eventually act as assistant to Kim or Ross or their more cognitively capable descendants.

This prospect causes conniptions among some consultants and accountants. It may not happen at the top for years — or ever, in complex lawsuits or tax audits. But aspiring partners should start honing their listening skills, just in case.

Wednesday, February 3, 2016

These four-foot-tall robots could change the way warehouse workers do their jobs

Tech Insider | Leanna Garfield | Feb 2, 2016

Locus Robotics

The Locus robot can zip around a clothing warehouse larger than the size of six football fields. It can also work for 24 hours, without a break for lunch or a salary.


The new bot, created by Locus Robotics, just launched in its first warehouse: a Devens, Massachusetts space owned by Quiet Logistics, a warehousing company that fills online orders for both small startups and megabrands like Zara and Bonobos. The robots transport items that have been picked off the shelves by humans, and bring them to the front of the warehouse to be sealed and delivered.

"We developed a system where the robots do all the walking," Locus Robotics CEO Bruce Welty tells Tech Insider. "As retailers continue to exceed expectation around next-day shipping, they're going to look to technology to help them provide an even faster turn-around."

The bots work alongside humans and do all the normal grunt work. Warehouse workers usually walk 12 to 16 miles each day. With the robots, they don't have to.

The robots now meet the human workers in the middle of the warehouse. As soon as someone completes an order online, the bot's system knows exactly where to go in the 275,000-square-foot warehouse.

Each bot (which doesn't look anything like a human) has a platform for arms and a two-foot-diameter base with wheels for feet. It zips around at about 4.5 mph, or the equivalent of a fast walk. If stray boxes or wires stand in its path, its vision technology can "see" in real-time to avoid them.

locus robot picking
Locus Robotics

Since the robots are able to move faster than humans without tiring, Welty says the system will boost warehouse productivity by up to 800%. The bots will also not be subject to human error, which means that they can get the order right nearly every time.

As a way to increase productivity and speed, many online retailers have been using robots, conveyor belts, and cranes to fulfill orders for the past decade. Amazon has exclusive rightsto Kiva Systems' robots, but Welty says Locus' robots are smaller, and more lightweight and versatile.

Unlike Kiva's bots, the Locus robots can be incorporated into the warehouses' existing infrastructure. The warehouse doesn't need to move any shelves or aisles. Locus can program the robot to navigate the warehouse exactly how it is.
locus robot 1Locus Robotics
"The robots are less industrial and easier to work side-by-side with humans," Welty says.

The robots won't replace any human jobs in Quiet Logistics' warehouse, at least for the time being. It will change the nature of many of their responsibilities, however. Since they don't need to roam the warehouse's aisles looking for the items, they can use that time for other parts of the distribution process. They can now do what they do best: add a personal touch.

As more retailers shift online and promise greater shipping speeds, there's a growing need to make sure that the orders are not only delivered on-time but also feel personalized. This can be as simple as stuffing boxes with pretty tissue paper or even handwritten notes from the brands — a job fit best for humans.

Tuesday, January 26, 2016

AI Politics: How An Artificial Intelligence Algorithm Can Write Political Speeches

By Menchie Mendoza, Tech Times | January 26, 5:43 AM



Soon, the speeches that we hear from political figures may be the product of an AI machine that has been specifically designed to write political discourses.


There seems to be a formula to writing speeches. For instance, political speeches sound similar and tend to have a standard format. Arguments in these political discourses seem to be repeated as well. Speeches also seem to use familiar phrases that show the speaker's certain political affiliation or ideology.

Valentin Kassarnig of the University of Massachusetts Amherst took all of these into consideration and made an AI machine to rival human speech writers.

"In this report we present a system that can generate political speeches for a desired political party," wrote Kassarnig. "Furthermore, the system allows to specify whether a speech should hold a supportive or opposing opinion."

Kassarnig said that in order to start training a machine-learning algorithm, he first built a database of around 4,000 political speech nuggets that he got from 53 Congressional floor debates in the United States. He gathered more than 50,000 sentences out of the speeches, with each sentence having 23 words on average. He also divided the speeches into categories depending on the political party (Democrat or Republican) and whether the speech is for or against a certain topic.

After trying various database analyzing techniques, Kassarnig decided to use a method based on the n-grams approach, which focuses on analyzing sequences of words or phrases.

First, he studied and tagged each word based on what part of speech it is (noun, adjective, verb, etc.).

Next, he turned to 6-grams and tried to determine all the words that can appear after the five previous ones, which include the probability of their appearance.

When these steps are accomplished, the process of creating speeches will automatically follow.

In his report "Political Speech Generation," Kassarnig also indicated the use of two underlying models from where he based the probabilities in the word sequencing. The language model handles the grammatical correctness of the speech while the topic model is aimed at achieving textual consistency.

He added that he also used both the manual and the automated approaches when evaluating the quality of AI machine-generated speeches. After performing an experimental evaluation, Kassarnig learned that in essence, generated speeches are good when it comes to correct grammar and the way the sentences would transition.

If you're curious to know whether a machine can actually generate good speeches, here is one example of a speech with a Democratic tone that was generated automatically by the AI algorithm.

"Mr. Speaker, for years, honest but unfortunate consumers have had the ability to plead their case to come under bankruptcy protection and have their reasonable and valid debts discharged. The way the system is supposed to work, the bankruptcy court evaluates various factors including income, assets and debt to determine what debts can be paid and how consumers can get back on their feet. Stand up for growth and opportunity. Pass this legislation."

While the AI algorithm shows huge potential in generating decent political speeches, Kassarnig is not liiting the potential of the algorithm to just politics. Instead, he suggests that the algorithm can also produce other types of texts, including news articles and blog posts.