By Eric Niiler
In the new techno-thriller “Upload,” a young computer scientist with a sketchy past and distrust of society decides to take the ultimate leap forward by scanning his brain and uploading his memories, personality and consciousness into a simulated world of his own making.
Raymond wants to live forever, controlling his environment and interactions with other humans as a god-like being.
The novel by author Mark McClelland is set in the Michigan of 2070 about the time that futurists like Ray Kurzweil predict that “singularity” will be reached, the moment when machine learning will surpass human intelligence. It’s not the first science-fiction tale to explore human-computer hybrids (see “What are Little Girls Made Of” in the first season of the original “Star Trek” series) or even the perils of virtual reality becoming too real (see the “Matrix” triology). But it does posit some questions that real-world researchers are just now tackling.
The European Union, for example, recently announced it was funding a $1.3 billion project to build a human brain on a silicon substrate. That’s about 1 1/2 cents per neuron. Swiss neuroscientist Henry Markham, who is behind the Human Brain Project, has already started work on building a simulated rat brain.
At the same time, getting machines to think more like humans is also progressing. Especially in the realm of human-computer interfaces. Researchers are figuring out ways for victims of paralysis to control their artificial hands with brainwaves or even walk with exoskeletons.
Both computers and the robots they control are getting smarter, according to Moshe Vardi, professor of computer science at Rice University.
“Are machines getting more and more powerful? Absolutely. It’s been going on since 1940. We are making progress, and for many people it will be a lifesaver. But we are very far from understanding how the brain works.”
Author McClelland, who still works as a software developer for a Chicago trading firm, says that he expects developments in computing and artificial intelligence to proceed much more quickly than similar advances in neuroscience.
“The biggest issue we face right now in modeling human intelligence is the sheer number of computations happening at the same time,” McClelland said. “The scale of interconnected calculation is out of reach today.”
Still McClelland said that if such a technology were developed, he would probably try it, just as his anti-hero Raymond.
“I’m not sure I would want to be the first one to do it,” he said. “There is tremendous appeal. The obvious benefit of having a longer lifespan, the possibility of living in a fantasy world for a period of time is certainly appealing. There’s an obvious risk of losing your philosophical grounding and becoming depressed and realizing your world doesn’t matter.”
For his part, researcher Vardi says that uploading the human brain and consciousness onto a computer is still in the realm of science fiction.
“It depends what you think about death,” Vardi said. “Some people have a hard time with this concept that you don’t exist anymore. This is one more attempt to overcome death. I don’t find it particular useful.”