[SOUND] Robot overlords, sentient computers.
I'm sorry Dave, I'm afraid I can't do that.
Armagheddon, this is what many people fear when artificial intelligence, or AI comes to mind.
[MUSIC]
I'm Martin Hitch and I'm the chief business officer of Bossa Nova Robotics.
I think AI is misunderstood.
We have this expectation that it's gonna take over.
And the reality is we're many, many, many years away from AI being sent here.
Instead, what we are seeing today are devices like Amazon's Alexa, Google's Home, or Apple's Siri.
These gadgets may sound sentient.
It's 61 degrees fahrenheit with mostly sunny skies.
But really, they are far from it.
Hi, I'm Robert Monroe, and I am the CTO of the great.
The AIs that we're building are There are AI's that we can smartly interact with.
So they're not optimized to be independently intelligent,they're optimised to be our guides, our systems, the extensions of our memory.
Aside from consumer electronics artificial intelligence Greatest promise may be in transforming entire industries.
The underlying technology making this happen is known in the tech community as supervised learning.
You have humans who are able to provide the training data for algorithms to them automate a task.
So for example, a self-driving car knows how to avoid a pedestrian, knows how to stay in the lane.
Because humans have sat down with images taken from the front of these cars and said, this is a human.
This is the lane which marks the edge of the road.
And it's these thousands and thousands of hours of human feedback which ultimately enables a self-driving car to be trained to automate this car.
[MUSIC]
Google's Waymo is on the forefront of self-driving car technology.
And is using supervised learning to drive it forward.
But, in fact, transportation is just one area that is booming in the field of artificial intelligence.
Another is healthcare.
Hi I'm Fabien Beckers, CEO and cofounder of Arterys.
Meet Arterys.
Their machines have been trained on thousands of patient images which gives physicians and radiologists better information for diagnosing cancer in the brain and lungs, and also detecting dangerous heart conditions.
[UNKNOWN] address.
Now, we have to be highly trained to be able to look at the organ.
We spent a lot of times [UNKNOWN] drawing contours by hand [UNKNOWN] the ventricles or the atrium, and that [UNKNOWN] of time.
And so what we have been working on is automating that and automating [UNKNOWN].
We trained the system on thousand images what we know exactly what the answer is [UNKNOWN] because the system can repurposed A suggestion to the position that would be very close to what he or she would have done.>>
[MUSIC]
Agriculture is also seeing a big change in the use of artificial intelligence.>> I'm [UNKNOWN] t. I'm the CEO of November technology.>> This is Blue River technologies.
See and spray machine Using artificial intelligence, the robot is able to detect which plant is a weed and which is a crop.
This is one of our 12-row machines, it goes to the field very fast, it's a very wide machine and back there, there's a camera that is looking down at the ground.
It looks at every single plant, 50 times per second, every camera that it has is looking at every plant and deciding.
Is this a crop?
Is this a weed?
And applying herbicides only to the weed and nowhere else and by doing that we can reduce about 95% of the herbicides being used.
We have a database that is growing really fast, now it's close to a million images that we've collected across many different farms, many different geographys.
That gives us the variability that the algorithms need in order to be trained and reliably detect which ones are weeds and which ones are crops.
From farming to the food that you shop for AI is impacting the whole supply chain.
At Bossonova Robotics, robots scan the aisle [UNKNOWN] of the supermarket aiding retailers in logging grocery inventory.
The robots that are designed to move autonomously through a store taking very high resolution pictures of the shelves, and we then use a very complex AI engine that extracts data from those shelves.
So, think about product data, label data, the layout of the items on the shelf.
Are they in stock, are they out of stock?
All that information happens in near real time.
[MUSIC]
The robots use computer vision to see the store shelves, and are trained on thousands of images to complete their task.
You could imagine now, you're starting to gather a very, very large data set We use that to train AI algorithms to identify specifics.
If we're successful, we blend it to the background and your shopping trip means you buy 100% of the items on your shopping list instead of 80% of them.
[MUSIC]
Of course, robots Sticking round the aisles of the supermarket, or tractors that can automatically graze crops, is just the beginning.
Experts believe a profound technological shift is under way.
[BLANK_AUDIO]
