The future is now
Welcome to the future. (Screencap from Love, Death + Robots)
Dr. Nando de Freitas, a lead researcher at Google's DeepMind, has declared "the game is over" when it comes to achieving artificial general intelligence (AGI) aka human-level intelligence.
Freitas took to Twitter in response to an op-ed by Tristan Greene following the release of DeepMind's Gato, in which Greene suggests that humanity may never achieve AGI and that it at least seems “like AGI won’t be happening in our lifetimes." He added, “Basically, Gato and GPT-3 are both robust AI systems, but neither of them are capable of general intelligence.”
But it appears that de Freitas, like OpenAI Chief Scientist Ilya Sutskever, believes otherwise. “Solving these scaling challenges is what will deliver AGI,” the DeepMind researcher tweeted, later adding that Sutskever “is right” to claim, quite controversially, that some neural networks may already be “slightly conscious.”
DeepMind itself hasn’t gone so far as to declare its new Gato multi-modal AI system capable of AGI, but given what one of its lead researchers is saying, it’s only a matter of time before Google declares that it’s going to be the first to achieve it.
Greene clapped back with another piece that basically stated, neither Gato, nor other AI models like DALL-E and GPT-3, are robust enough for unrestrained public consumption. Each of them require hard filters to keep them from tilting toward bias and, worse, none of them are capable of outputting solid results consistently. And not just because we haven’t figured out the secret sauce to coding AGI, but also because human problems are often complex and they don’t always have a single solution.
Fly me to the moon
After a failed attempt back in 2019, the new Boeing passenger spacecraft, the CST-100 Starliner, successfully docked itself to the International Space Station—demonstrating that the vehicle can potentially bring humans to the ISS in the future. It’s a crucial capability that Starliner has finally validated in space after years of delays and failures.
Starliner is in the midst of a key test flight for NASA called OFT-2, for Orbital Flight Test-2. The capsule, developed by Boeing for NASA’s Commercial Crew Program, was made to transport NASA’s astronauts to and from the space station. But before anyone climbs on board, NASA tasked the global aerospace company with conducting an uncrewed flight demonstration of Starliner to show that the capsule can hit all of the major milestones it’ll need to hit when the spacecraft begins carrying passengers.
Starliner has finally shown what it was designed to do. “Boeing Starliner spacecraft completes its historic first docking to the International Space Station opening a new avenue of access for crews to the orbiting laboratory,” Steve Siceloff, a communications representative for Boeing, said during the livestream of the docking.
It’s happening. Artificial intelligence is changing the world—and our lives—as we know it. Evidently, it brings with it advantages and risks…and visions of The Terminator—Arnold Schwarzenegger’s half human half cyborg face holding his gun to your head.
Another question that often comes up when talking about AI is, “will there still be anything in store for us humans?” Maybe we should just fly to the moon.
