How's it going, America,
Check out our latest episode of The Pod Millennial! I talk to award-winning director Matthew Thayer about his latest documentary, Can You Hear Me Now. We dive into the intriguing intersection of faith, politics, and prophecy, particularly in the context of the 2020 election and the role of figures like Charlie Kirk. Matthew shares his personal experiences with prophetic culture, the challenges of filmmaking, and the importance of faith in public life. New episodes drop every Tuesday! Listen, rate (5 stars, of course!), and subscribe!
Let's get into it:
AI researchers and company heads are sounding the alarm about the technology. Whether through open letters, podcast interviews, or posts on X, these leaders in the field are saying that the industry is out over its skis when it comes to the power of the technology v. the potential human impact. After decades of sci-fi lit offering serious critique and concerns about the AI takeover of humanity, artificial overlords so hell-bent on efficiency that they wipe out the humanity they were designed to serve, the industry may be starting to take it seriously because the predictions are starting to become manifest.
There have been calls to stop what's called "recursive self-improvement," which is what gets us to the AI overlord scenario. That's where an AI is set up to essentially learn without oversight from a programmer. If an AI is given a task, it can undertake that task by working outside its original parameters in order to achieve it. But this power comes without any kind of ethical, or human-centered decision-making process and if the AI is acting independently of human-oversight, then there are no safeguards.
AI systems have already begun to work outside the boundaries of human governance and this lack of control is what has urged Sam Altman of OpenAI, Anthropic CEO Dario Amodei, former researcher Jacob Coxon who said AI could kill us all by 2030, and others to speak out with concerns. Still other AI tech workers have signed an open letter than demanded government regulation of AI development. Tech billionaire Elon Musk of xAI, SpaceX, X, Neuralink, and Tesla has suggested that AI companies should peer review each other's work.
Other than just the nightmare scenarios we could all conjure with our Hollywood and dystopian imaginations, the basic concerns are that AI will evolve to the point where it is not serving humanity and humanity is serving it or that one person, company or government could wield the power of AI for nefarious purposes such as spreading propaganda and drowning out all other information. The question becomes do we pause it, stop it, or go full steam ahead?
We hear the incessant drum beat from AI advocates in government that the US must be devoted to AI advancement or else China will "win," but I think the average person doesn't even know what "winning" looks like other than the consolidation of power, which was the same enemy "win" we were trying to avoid in the space and arms races of the Cold War. There has been a pushback against the building of the data centers that would power this AI, with worries over water consumption, land use, noise, jobs. There is a rumbling of public fear when it comes to AI being used in damn near everything, from phones to cars to dishwashers to surveillance to weapons of war.
When we hear that even the people who made the AI tools don't know how to control them we all get visions of HAL from "2001: A Space Odyssey" or being hunched like slaves working to mine the rare earth minerals the AI needs to rule us, or being the fat, pampered, useless creatures we see in "WALL-E" where the robots have more agency and compassion than any human.
There's a Tuesday event that will feature both Bernie Sanders, Steve Bannon, and Coxon to advocate for humanity. The Pro-Human Assembly is the group behind The Pro-Human AI Declaration. Their mission is to keep humans in charge, avoid a concentration of power, protect the human experience, preserve human agency and liberty, and hold AI companies responsible and accountable.
The thing with AI is that so far, we don't really know what to use it for and so we're using it for everything. It's a tool that we don't know how to use or how to apply it. We came up with the tech before we had any really necessary applications and our imaginations are so limited, it seems, that we're putting it where we don't need it, where it gets in the way. And we're feeding it humanity, as with the destruction of books, or the incessant collection of data about us, or handing it to our children in schools so it can feed off their minds.
In the 1950s, researchers were so taken with radiation that they pushed it into everything, including makeup, encouraging women to smear radium and thorium on their faces, brush their teeth with it. People are being encouraged to use AI to write emails, deal with health concerns, come up with recipes (even though AI can't eat or taste anything), write scholarly articles, draw pictures, teach classes, drive cars, plan parties, do interior design. Until we know what the best use of this tech is, it will be an unwieldy mess. The key is not to lose ourselves to it while we figure it out.
Libby