:00:00. > :00:00.Now on BBC News it is time for Click.
:00:00. > :00:07.This week, mapping the poorest parts of the world.
:00:08. > :00:37.And, hands up who is not flying the plane?
:00:38. > :00:46.There are things happening in artificial intelligence right
:00:47. > :00:51.now that will fundamentally change our world.
:00:52. > :00:56.Soon, machines will learn to do our jobs.
:00:57. > :01:08.And at that point, things get very interesting.
:01:09. > :01:11.We'll talk more about the consequences of an automated
:01:12. > :01:14.society in a few minutes but, after thinking, walking and driving,
:01:15. > :01:24.have you ever wondered how hard it would be for a computer to fly?
:01:25. > :01:28.I'm not talking about drones that can fly between points, follow
:01:29. > :01:35.I'm talking about aircraft that could intelligently
:01:36. > :01:41.decide on a flight plan, just as a human would.
:01:42. > :01:48.And that is what Mark has been hunting down.
:01:49. > :01:52.Here at BAE Systems in Woolton in Lancashire, they are used
:01:53. > :02:00.They built and have tested Eurofighter Typhoons here.
:02:01. > :02:04.Today, however, I am going to take a flight in an aircraft that is much
:02:05. > :02:12.This is a Jetstream 31, a small passenger aircraft.
:02:13. > :02:16.It's a design from the 1980s but it's currently used by BAE
:02:17. > :02:19.Systems as a flying test-bed for technology which could lead
:02:20. > :02:27.Maureen McCue is head of research here.
:02:28. > :02:32.Very well flown and understood aircraft from the outside,
:02:33. > :02:37.but on the inside, it's filled with the latest technology.
:02:38. > :02:43.That technology will eventually allow this aircraft to fly
:02:44. > :02:47.Today, they are testing the plane's ability to detect and avoid
:02:48. > :02:52.clouds as well as testing its satellite communication systems.
:02:53. > :02:55.But take-off and landing will still be handled by human
:02:56. > :02:57.pilots and the plane will be remotely controlled
:02:58. > :03:01.How does this fit into the autonomous equation?
:03:02. > :03:03.At the moment, it's effectively a remotely controlled aircraft.
:03:04. > :03:06.It is, and really, with autonomous operations,
:03:07. > :03:09.you need to progressively expand the boundary.
:03:10. > :03:13.You can't start with such a big bang right out at the full range
:03:14. > :03:23.This humble looking outbuilding houses the grand station.
:03:24. > :03:27.Here, a pilot will remotely fly the plane and he can ensure it
:03:28. > :03:35.will react to instructions from air traffic control.
:03:36. > :03:39.I would expect to see a joystick and images coming
:03:40. > :03:42.through from the cockpit by you're not going to be flying
:03:43. > :03:47.No, everything is through the numbers that you can see there.
:03:48. > :03:53.These flights are taking place in uncongested airspace.
:03:54. > :03:59.Today, we will be flying over the Irish Sea.
:04:00. > :04:10.To help fly itself, this aircraft uses data from satellites
:04:11. > :04:13.as well as identifying radio signals broadcast by other aeroplanes, so it
:04:14. > :04:20.It is also fitted with a camera that can see other air
:04:21. > :04:24.users, even if there are not warning signals.
:04:25. > :04:26.So right now, the pilots aren't actually flying the aircraft.
:04:27. > :04:34.From that 15-year-old Dell laptop that's probably Windows XP.
:04:35. > :04:38.At this point, the aircraft is flying autonomously with a human
:04:39. > :04:46.Handing control of the computing over to the autopilot in the back
:04:47. > :04:51.and once established on the route, I can hand the computing
:04:52. > :04:57.over to Clive by the satellite on the ground.
:04:58. > :05:00.So that little shed-like building we were in earlier with Clive
:05:01. > :05:04.in front of the computer, he is now flying the aircraft.
:05:05. > :05:09.Over the course of the testing of this aircraft, it's going to have
:05:10. > :05:12.to perform a variety of different, complex tasks.
:05:13. > :05:15.For instance, it's going to have to recognise and avoid bad weather.
:05:16. > :05:19.Not just weather, but other aeroplanes, too.
:05:20. > :05:22.It will eventually be able to select a safe landing spot
:05:23. > :05:30.Today, we can't really test its weather detection
:05:31. > :05:32.abilities though as, unusually for the UK,
:05:33. > :05:38.BAE suggests that autonomous aircraft could be used
:05:39. > :05:42.to perform dirty, dangerous or repetitive tasks.
:05:43. > :05:44.But could this technology be introduced into
:05:45. > :05:50.At the moment all commercial aircraft have a set number of crew.
:05:51. > :05:53.There are programmes in existence looking at how you reduce crew
:05:54. > :05:58.either planned from the outset or, in the case of an emergency,
:05:59. > :06:02.you've got the autonomous system as a fallback so that you can
:06:03. > :06:05.still have perhaps a two-crewed aircraft on a certain length
:06:06. > :06:09.of flight but one of the crew happens to be an autonomous helper
:06:10. > :06:19.But what happens when things go wrong?
:06:20. > :06:31.While aerospace manufacturers are exploring the possibilities
:06:32. > :06:34.of fewer cockpit crew, what do commercial pilots think?
:06:35. > :06:39.To find out, we paid the British Airline Pilots Association a visit.
:06:40. > :06:42.Many decades of looking at aviation has brought us to the position
:06:43. > :06:46.where we have come to the conclusion it's best to have two
:06:47. > :06:49.pilots in the cockpit, because if you reduce that to one,
:06:50. > :06:52.the problem you've got then is you've got no one to cross check
:06:53. > :06:55.Take for example, the miracle on the Hudson.
:06:56. > :07:06.When the aircraft lost both its engines, the pilots had
:07:07. > :07:09.to have a discussion and they decided their only course
:07:10. > :07:13.No computer can be programmed to do that.
:07:14. > :07:16.The flight testing of autonomous aircraft continues but the debate
:07:17. > :07:19.about regulating them and how they are going to be used has
:07:20. > :07:23.That was Mark and this is Tim Harford, columnist
:07:24. > :07:26.for the Financial Times and Tim, you've written a book about how
:07:27. > :07:32.the systems that we now rely on can sometimes backfire.
:07:33. > :07:36.What do you make of the idea of planes that might only need one
:07:37. > :07:46.And of course, autopilots have made planes safer
:07:47. > :07:48.but what worries me is, what happens when
:07:49. > :07:51.No system is perfect, including a system where
:07:52. > :07:55.I guess when it goes wrong, it has two hand back
:07:56. > :08:00.Autopilot hands back to the human in the cockpit but then what?
:08:01. > :08:03.The human is out of practice, the human is not used to flying
:08:04. > :08:06.the plane and because the autopilot has failed, it's probably
:08:07. > :08:11.There is a worrying example of this a few years ago.
:08:12. > :08:14.An Air France crash over the Atlantic Ocean.
:08:15. > :08:18.The plane was flying quite high above a storm.
:08:19. > :08:23.The autopilot disconnected and the pilots just weren't used
:08:24. > :08:32.They were only used to operating the plane on take-off and landing
:08:33. > :08:35.and they flew a perfectly good plane into the Atlantic Ocean
:08:36. > :08:37.because they were confused about what was happening.
:08:38. > :08:40.They killed everybody on board, an absolute tragedy, and this
:08:41. > :08:45.The autopilots are normally so safe, so reliable,
:08:46. > :08:47.that when they fail, the pilots find
:08:48. > :08:53.I guess the next question is, what about autonomous cars?
:08:54. > :08:57.We have been talking about how they will blissfully drive us around
:08:58. > :09:02.I suppose for the foreseeable future, they won't be good enough
:09:03. > :09:10.I guess they'll never be 100% reliable but the model
:09:11. > :09:13.where if it's confused, it hands back to the human,
:09:14. > :09:22.You are there with your bagel, your coffee, your newspaper.
:09:23. > :09:25.You look up, there's a bus coming towards you, and the car goes,
:09:26. > :09:28.autopilot disengaged, human take control, it clearly not
:09:29. > :09:32.What makes more sense is for the human to be driving
:09:33. > :09:36.and for the computer to be watching out for a dangerous situation,
:09:37. > :09:38.for the computer to take over if there's a problem.
:09:39. > :09:43.Humans get bored, get distracted, lose their skills.
:09:44. > :09:49.None of these things happened to computers.
:09:50. > :09:53.I guess we are in an extended period of time before the far future
:09:54. > :09:55.happens and computers drive and fly us.
:09:56. > :09:58.We've got possibly decades if not a century of being in this
:09:59. > :10:01.interim period where, if there is a problem,
:10:02. > :10:04.we are going to end up blaming them for this really unusual,
:10:05. > :10:07.weird crash that a human wouldn't have made.
:10:08. > :10:10.Yes, and I think a glimpse of that is where we are asking
:10:11. > :10:14.the computers to make a decision not about planes or cars but about,
:10:15. > :10:18.for example, who'll get a promotion or who gets a special deal in a shop
:10:19. > :10:21.or who gets arrested for shoplifting because the computer
:10:22. > :10:26.We are already asking computers to make this sort of decision
:10:27. > :10:30.and the lesson of the paradox of automation is that we need to be
:10:31. > :10:33.much more savvy about the fact that computers do make mistakes
:10:34. > :10:45.Hello and welcome to The Week In Tech.
:10:46. > :10:48.It was the week that inventor Haiyan Zhang developed a smart
:10:49. > :10:50.wristband to help people with Parkinson's disease
:10:51. > :10:56.The device's in-built motors vibrate to distract
:10:57. > :11:02.It was also the week that we discovered queueing
:11:03. > :11:04.at the shops and using those beyond infuriating self-service
:11:05. > :11:08.checkouts could soon be a thing of the past.
:11:09. > :11:11.Amazon has unveiled a sci-fi store in Seattle that uses your smartphone
:11:12. > :11:16.and advanced technologies like deep learning, computer vision and sensor
:11:17. > :11:20.fusion to automatically detect when products are taken
:11:21. > :11:26.When you're done, you can simply trot off and then wait
:11:27. > :11:30.for that gargantuan virtual receipt to follow.
:11:31. > :11:32.And if you feel like you're forever stuck in traffic,
:11:33. > :11:36.Audi has rolled out an update to make every second count.
:11:37. > :11:39.Its new traffic light information feature tells drivers exactly how
:11:40. > :11:42.long they'll have to wait behind a red signal before it turns
:11:43. > :11:47.green and the length of time it will stay green.
:11:48. > :11:50.It works by connecting directly to the city's
:11:51. > :11:56.And finally, robotic research has reached new heights, literally.
:11:57. > :12:04.This hopping mad bot developed at UC Berkley cannot only jump a meter off
:12:05. > :12:07.the ground but can then again jump off objects to reach
:12:08. > :12:12.Inspired by the agility of bushbabies, researchers hope
:12:13. > :12:21.it can one day identify jumping spots for itself.
:12:22. > :12:30.Which these days isn't guaranteed to be true.
:12:31. > :12:33.In the run-up to the US election, for example, the Speaker
:12:34. > :12:37.of the House of Representatives did not get naked, the Pope did not
:12:38. > :12:40.endorse Donald Trump and he did not win the popular vote,
:12:41. > :12:45.but these stories, from websites posing as real news sites,
:12:46. > :12:51.Of course, it doesn't help that in 2016, the real news sounds
:12:52. > :12:57.But anyway, it's made events like the Trust Hack here in London
:12:58. > :13:06.Here, journalists and technologists from large news organisations
:13:07. > :13:11.are workshopping ways to help readers tell the difference
:13:12. > :13:13.between well-researched journalism, propaganda, advertising,
:13:14. > :13:20.The thought is to provide images like icons back-up materials
:13:21. > :13:24.that the public could see connected to a piece of news and then it
:13:25. > :13:27.would send a signal back to the news distribution platform like Google
:13:28. > :13:31.or Twitter so that they can identify quality news out of the fake news
:13:32. > :13:48.There are already projects afoot to try to flag up stories on sites
:13:49. > :13:50.known to generate fake news like this plug-in but the ideas
:13:51. > :13:53.here are not about blacklisting sites or producing automated
:13:54. > :13:56.Both would be massive undertakings and would themselves provoke cries
:13:57. > :14:00.This is more about letting news organisations prove to their readers
:14:01. > :14:05.The journals of the Washington Post, I work with some amazing people.
:14:06. > :14:08.We produce really great stuff and they remain really committed
:14:09. > :14:12.The tools that we are building here are just a way for us
:14:13. > :14:16.to communicate that we are putting in the effort, where our stuff
:14:17. > :14:18.is coming from, who we are talking to.
:14:19. > :14:29.We are trying to create something that would easily allow audiences
:14:30. > :14:32.to verify for themselves what sources we have used.
:14:33. > :14:36.You'd be able to click and see, who did we talk to, and you'd
:14:37. > :14:42.Readers want to feel like journals are being held accountable to them
:14:43. > :14:49.Other ideas here include ways to fight information bias
:14:50. > :14:52.by surfacing articles that support the opposite side of an argument
:14:53. > :14:55.or to look at the likelihood of truth by finding similar articles
:14:56. > :14:58.Many think the reputation of the journalists themselves plays
:14:59. > :15:00.a big part in the trustworthiness of reporting.
:15:01. > :15:02.Italian newspaper La Stampa is suggesting a system
:15:03. > :15:06.where an author is assigned a unique identifier that shows their piece
:15:07. > :15:17.The best ideas won a small prize at the end of the day but it's
:15:18. > :15:22.The hope is that this part-Google funded initiative may lead
:15:23. > :15:25.to a system that helps news outlet stories rank high up on search
:15:26. > :15:38.That said, I can't help wondering if that is actually something that
:15:39. > :15:40.platforms like Google and Facebook really want.
:15:41. > :15:43.Do you think they care what it is they serve to us or do
:15:44. > :15:47.you think, really, as long as you click on it, that's
:15:48. > :15:51.How do we know the motivations of any company?
:15:52. > :15:54.These companies make their money through people clicking
:15:55. > :15:57.on the adverts, so do you think any of the large companies care
:15:58. > :16:07.Based on my conversations with them, I think they do.
:16:08. > :16:10.The argument would be if they start being perceived as not caring
:16:11. > :16:12.about the information that's sent out there,
:16:13. > :16:25.Yes, it may be that the truth will out, not because of a desire
:16:26. > :16:28.for the facts, but because everyone, readers and news aggregators,
:16:29. > :16:48.More and more people are shopping online but still at this time
:16:49. > :16:51.of year, the high street seems pretty chaotic and the retailers
:16:52. > :17:01.So they are trying to create some more engaging experiences.
:17:02. > :17:05.But do they help us or are they just a distraction?
:17:06. > :17:08.Here in London's Covent Garden, 140 shops and restaurants are taking
:17:09. > :17:13.part in creating one huge augmented reality experience.
:17:14. > :17:21.With the help of AR app Blipper, things come to life.
:17:22. > :17:24.It may not have created the personalised shopping experience
:17:25. > :17:27.I dreamt up, but there were some promotional offers presented
:17:28. > :17:32.as virtual Christmas presents almost around the tree.
:17:33. > :17:35.A reindeer hunt and a giant reindeer you can take a selfie with,
:17:36. > :17:41.I have to say, it wasn't quite as cutting-edge as I'd hoped
:17:42. > :17:45.but I suppose it's a bit of light-hearted fun.
:17:46. > :17:48.Rather more purposefully, the signs in windows can be scanned
:17:49. > :17:50.using image recognition, taking you to online content,
:17:51. > :17:54.partly the sort of stuff you'd be able to look at from your sofa
:17:55. > :18:01.Then came our trip to London's Westfield where augmented
:18:02. > :18:09.We've seen technology like this before but now it's actually
:18:10. > :18:12.on the shop floor here at Charlotte Tilbury.
:18:13. > :18:14.This is what's known as the Magic Mirror and this
:18:15. > :18:19.You choose a lipstick and in real-time, you will see
:18:20. > :18:28.Bright red lips, although it doesn't seem to have any around the edges,
:18:29. > :18:38.I've tried the Rimmel app that does something similar on your phone
:18:39. > :18:42.and you can buy things through it, but here you can do it in the store
:18:43. > :18:46.with assistants all around and a whole shop of products that
:18:47. > :18:49.you can test, smell, and after you see what your face
:18:50. > :18:52.looks like on here, you might want to have a go to check that
:18:53. > :18:57.Meanwhile, here at this eBay event, they are taking things
:18:58. > :19:02.The data on this screen represents what is apparently visitors
:19:03. > :19:06.Using what they call facial coding, the camera looks for reactions
:19:07. > :19:09.which these guys reckon you have when you do online shopping.
:19:10. > :19:12.Nice but don't know who I'd give it to.
:19:13. > :19:27.And I've been told to overact my reactions.
:19:28. > :19:32.Whilst my results bore absolutely no correlation to what I'd liked,
:19:33. > :19:35.maybe they were the ones I contorted my face to the most.
:19:36. > :19:38.Maybe it would have worked better if I'd reacted more naturally,
:19:39. > :19:41.although I struggle to imagine that my face would have shown anything.
:19:42. > :19:44.It left me wondering whether eBay could be developing this if it
:19:45. > :19:47.worked as something more permanent to assess our feelings when online
:19:48. > :19:57.Either way, don't expect me to look too excited about it.
:19:58. > :20:02.Not that I'm sure the tech would have even noticed.
:20:03. > :20:05.Now, in 2015, members of the United Nations adopted a set
:20:06. > :20:10.Number one on that list is to end poverty and to achieve that goal,
:20:11. > :20:13.you first need to work out where poverty exists and how
:20:14. > :20:23.We met up with some scientists at Stamford who have that
:20:24. > :20:26.Marshall Burke is a professor of earth systems science
:20:27. > :20:30.at Stanford University but he spends much of his time in Africa
:20:31. > :20:38.The way this is done is to elicit from the household a listing
:20:39. > :20:40.of everything they've consumed in the last week,
:20:41. > :20:47.So literally everything they've consumed.
:20:48. > :20:51.Every single thing and the value of that item and then you add up
:20:52. > :20:53.all these items for every single person in the household.
:20:54. > :20:56.This can take hours and hours just for one single household.
:20:57. > :21:00.Then you have to do this for thousands of households to get
:21:01. > :21:06.It's painstaking work but Burke has teamed with computer science
:21:07. > :21:12.Using machine learning to predict poverty data
:21:13. > :21:18.But to find out whether the people living in those areas are rich
:21:19. > :21:20.or poor, the researchers used a process called transferred
:21:21. > :21:26.learning and this image of the Earth at night.
:21:27. > :21:30.The parts of the world that are lit up are typically the wealthier parts
:21:31. > :21:34.So basically we use the lower resolution night-time images to help
:21:35. > :21:37.us figure out what in the really high resolution daytime images
:21:38. > :21:40.we should should be using and then we use that to predict poverty
:21:41. > :21:52.Between 300 and 400,000 images were used to train the algorithm.
:21:53. > :21:53.The algorithm will figure out what's important,
:21:54. > :21:58.So some of the things it finds are things that
:21:59. > :22:00.you or I would recognise, things like roads,
:22:01. > :22:03.Based on those features, the algorithm can predict
:22:04. > :22:10.Things like refrigerators, cars, the sum of all those assets.
:22:11. > :22:13.It can also be used to predict incomes.
:22:14. > :22:15.These poverty maps show the team's findings.
:22:16. > :22:17.In areas marked red, people spend as little
:22:18. > :22:21.In green regions like Uganda's capital, Kampala, they spent
:22:22. > :22:30.We are providing a very cheap and scalable alternatives
:22:31. > :22:31.to traditional means of data collection.
:22:32. > :22:34.Traditionally, you have to send people out into the field
:22:35. > :22:36.with clipboards, the surveys aren't always accurate,
:22:37. > :22:40.Like for example lots of governments where they are underperforming,
:22:41. > :22:45.All we need to make our predictions are satellite images.
:22:46. > :22:48.But can you draw conclusions about the economic well-being
:22:49. > :22:50.of communities in Africa when you're thousands of miles away,
:22:51. > :22:53.sitting at a laptop in an office at Stamford?
:22:54. > :22:59.We actually have really good survey information in a few locations.
:23:00. > :23:02.We can use the satellite imagery to make a prediction about poverty
:23:03. > :23:05.and then we can compare that to what the survey says was actually
:23:06. > :23:11.So we used a couple of the really good surveys we had to validate
:23:12. > :23:20.To be a truly useful tool though, the algorithm needs an upgrade.
:23:21. > :23:24.We would also like to use historical imagery so maybe we can figure out
:23:25. > :23:27.how poverty dynamics work overtime and even give us the chance
:23:28. > :23:31.of predicting what's going to happen in the future.
:23:32. > :23:35.If you can pinpoint poverty on a map, aid could be distributed
:23:36. > :23:37.more evenly, policies could be more effective.
:23:38. > :23:45.A picture may be worth a thousand words but combining that picture
:23:46. > :23:52.with artificial intelligence could make a world of difference.
:23:53. > :23:54.That was Sumi and that's it for this week.
:23:55. > :23:57.You can follow us on Twitter @BBCClick for backstage fun
:23:58. > :24:00.and photos and extra technology news throughout the week.
:24:01. > :24:31.Thanks for watching and we'll see you soon.
:24:32. > :24:34.Well, it's still very mild and murky out there.