By Peter Weddle, Author Circa 2118: What Humans Will Do When Machines Take Over
There’s a newly extinct species on planet Earth. It’s the genus known as AI-naysayers. You probably heard them opining at conferences that there is no such thing as real artificial intelligence. You may have read their blogs and reports howling that AI is just a lot of hype. They were everywhere, and now they aren’t. They’ve joined the ranks of the dodo bird and those who proclaimed that job boards are dead.
What has caused their disappearance? Reality. Here’s just a sample.
Fei-Fei Li, the director or Stanford’s Artificial Intelligence Lab, has launched a program call AI4ALL that will introduce high school students to artificial intelligence so that they are better able “to think about their future participation in developing and guiding this technology.”
A computer called Spiking Neural Network Architecture or SpiNNaker for short has now come online. It’s the first-ever computer that’s able to complete more than 200 million actions per second, and it does so by emitting spikes of electric energy in just the same way the human brain does.
A neural network named Shelly after the author of Frankenstein has been taught how to write horror stories. Today, it both writes its own creations and collaborates with humans on co-authored novels. Look out James Patterson.
Google DeepMind has now achieved “near-human performance” in reading CT scans at the University College London Hospital. Why is that important? As the research team put it, “Increasing demands for and shortages of trained staff place a heavy burden on healthcare systems, which can lead to long delays for patients as radiotherapy is planned.” In other words, in the War for Talent, machines will be the victors.
Embedding from Language Models, which is also known as ELMo, has demonstrated the ability of AI systems to read and comprehend unstructured data and correctly answer SAT-like questions. This capability is the key to developing machines that can learn on their own and no longer need humans to program them.
A portrait painted by an AI machine at the art collective Obvious in Paris was given the title Edmond De Belamy and sold at Christies Auction House for a cool $400,000. By most accounts, neither Picasso nor Banksy have much to worry about, but the development is yet more proof that machines can learn how to be creative.
IBM has developed a way to build artificial synapses so that a computer can “think” the way humans do by passing information via electrical impulses between neurons. The process makes the time and effort required to teach machines how to do specific human tasks 100 times more efficient.
Accenture has been granted a U.S. patent for its ZBx AI platform, which uses both AI and machine learning to categorize financial transactions quickly, enabling faster and more accurate spend analysis. One fan described it as “the liberation of the finance professional from the tyranny of the spreadsheet.”
The AI system Flow Machine created the melody for lyrics that resulted in a pop song called “Daddy’s Car.” As one critic noted, it sounded “remarkably like a tune composed by the Beatles in the 1960s.” If it sells even half as many records as they did, it’ll be the first multimillionaire machine.
Not to be outdone, a high school student – that’s right, a kid with absolutely no formal college training – taught an AI machine to rap like Kanye West. At first, the machine simply mimicked the artist, but now, it’s generating lyrics on its own. Today’s American Music Awards may soon have a competitor – the American Machine Awards.
All of these and many other developments notwithstanding, we are still in the very early days of the AI Age. There’s clearly much work still to be done not only in building out the technology’s capabilities, but in understanding and managing its impact on individuals – especially working men and women – and our social, political and other institutions. What has been accomplished, however, is enough real progress to eradicate an invasive species – the AI-naysayer. Good riddance.