An AI Update
AI fears… AI hype…
An AI Update
AI fears… AI hype…
Let’s take a look at the state of AI, which some, like Yuval Noah Harari have suggested we call “Alien Intelligence” because there will likely come a time in the not-too-distant future where AI makes surprising and unpredictable choices, since, well, it’s not like us. It is designed by us and some there’s a certain degree of lock-in when it comes to the best and the worst parts of our current age. We still have rampant sexism, racism, homophobia, and plenty of other deep-rooted prejudices in our [Western] society. Part of the reason our Big Tech feels the need to race against China is because of these lock-in related concerns.
I certainly cannot make the argument that my knowledge is cutting edge given I’m by no means an expert in the field. That being said, I’ve been paying close attention. AI presents a massive threat against our society, our way of life, our future, and it feels essential to try to prepare for what is on the way. It’s arriving fast, too fast, and as disenfranchised members of society we don’t have much say in this.
*
Not long ago…
I made the terrible terrible mistake of reading ‘AI 2027’ at night before bed…
I was in a state of despair for about 48 hours after reading AI 2027… so I’m not exactly suggesting you read this.
AI 2027 is… a story. It’s near-future sci-fi vision in the vein of Black Mirror. But it sounds real and all too likely and, in turn, very very scary. I’ve read some counter arguments including from AI chatbots. The chatbots put the likelihood of future AI takeover at low-moderate risk (5-10%). That’s not entirely helpful (or comforting) from a source that, although not yet sentient so far as we know, might have a reason for slight bias.
+
AI 2027: Media, Reactions, Criticism
+
Update: The AI Futures Project published an update on June 28th:
‘What you can do about AI 2027’
*
Some other interesting AI-related discussions involve debate as to how fast tech and change will accelerate over the next decade (or less). This involves some complexities involving energy, compute, bottlenecks, and other concerns that I admittedly do not fully understand.
‘Will MacAskill on AI causing a “century in a decade” — and how we're completely unprepared’ (80,000 Hours)
Note: I read the highlights. These discussions are often 3+ hours… longwinded.
My initial response:
Probably nothing.
In all seriousness, the Ethical Altruism and moral philosophical positions take a defensive posture that is, well, a bit extreme, in a kind of counterpoint to the AI technocrat/technologist hype machine.
But then I read AI 2027. And, since then, other predictions.
We can sidestep bothsidesism by acknowledging that there are more than two sides to this conversation.
To be clear, insiders are concerned about the impact of AI, not just naysayers or those who think AI is overhyped.
Luckily, we do have forces such as the Center for Humane Technology fighting for digital citizenry rights and trying to save us from ourselves (that is, the human condition and our over-interest in new frontiers).
*
Tech guru Jaron Lanier: ‘The danger isn’t that AI destroys us. It’s that it drives us insane’ (Guardian)
*
The Techno-Optimist Manifesto by Andreessen Horowitz
Yeah, so, some of this is pretty unhinged…
*
‘Ninety-Five Theses on AI’ by Samuel Hammond
My understanding is this is widely considered an important document in the AI community.
Not that long of a read. There are some points made in jargon-y in-group language that require a bit of digging to unpack.
*
This part isn’t scary so much as useful to know… sources matter.
‘AI Platform Citation Patterns: How ChatGPT, Google AI Overviews, and Perplexity Source Information’
*
Outlook not so good.
But this isn’t really accurate.
John Warner gets deep into the discussion about how AI cannot write and cannot teach. That’s not what it’s doing when it takes inputs and hands you outputs.
*
Seth Godin – Productivity, AI and pushback
“Either you work for an AI or AI works for you.”
*
What Would Happen If the Chatbots Broke Free of Their Masters? (Literary Hub)
Best part about this informative article? The humor.
Very little humor in typical conversations and articles about AI. We could all use a bit of levity.
*
Looking back…
This may well already be outdated:
All the Things People Use AI for in 2025
A reminder of how fast AI tech is transforming the state of play.
*
Over time… I’ve shared quite a lot of my explorations into what’s going on in AI in SC Weekly posts.
What follows are just some examples I’ve shared over recent months.
+
Is AI Productivity Worth Our Humanity? with Prof. Michael Sandel (Your Undivided Attention)
Sandel is really interesting. I’m planning to learn more about his positions.
Tristan Harris is one of my favorite people. Harris broke away from Tech Bro culture and formed the Center for Humane Technology. Tristan, if you’re reading this, let me know what I can do to help the cause.
+
The Narrow Path: Sam Hammond on AI, Institutions, and the Fragile Future
(Your Undivided Attention)
“The race to develop ever-more-powerful AI is creating an unstable dynamic. It could lead us toward either dystopian centralized control or uncontrollable chaos. But there's a third option: a narrow path where technological power is matched with responsibility at every step.”
+
‘How to Think About Writing in the Age of AI, with John Warner’
I’m currently reading John Warner’s book ‘More Than Words: How to Think About Writing in the Age of AI’. Often, I’ve said that listening to the podcast is sufficient for many nonfiction books as the books themselves are often protracted essays fluffed up to fill space. Further, many of the more scholarly nonfiction works contain a good deal of material that is for field-specific insiders and not lay readers. In any case, Warner’s book is an example where I’d suggest the book has much to offer not covered in the interviews. A text worthy of tactile engagement.
It put my mind a bit at ease regarding generative AI and is a reminder that there are good minds at work reflecting on how to improve the educational system so that the students of today are better prepared for a very different looking near-future employment and workplace ecosystem.
+
"Just know, however, that if you ever get caught submitting AI, your name has a high probability of being forever black-listed with that publication and that editor. Speaking from experience."
Written By a Human or AI? (FundsforWriters)
+
“[…] As AI-generated content becomes more prevalent, a tipping point may be reached where people stop assuming everything is posted by a human and start thinking it’s from a bot. Even if humans still make up most of the internet, the default assumption will be that everything is AI until proven otherwise.”
How Do I Prove That I'm Human? (Plagiarism Today)
+
On AI “art”…
“It’s not a great feeling. It’s like a backhand to the face.”
+
Some AI humor…
‘Opinion: A.I. Could Put Artists, Actors, And Writers Out Of Jobs, But It Could Do Bad Things As Well’
+
Rethinking School in the Age of AI
+
Hillary Clinton? AI thought leader!? Yes indeed.
Hillary Rodham Clinton with Reid Hoffman — Superagency: What Could Possibly Go Right with Our AI Future
"Part of the thing with human beings is that we handle transitions badly." - Reid Hoffman
An important takeaway is that what’s coming is inevitable (with regards to AI) and it’s important that everyone take steps to educate themselves (at least a little).
I’ve read about this from numerous perspectives (really all four that Hoffman has… invented…)
AI cautious, doomers, gloomers, bloomers, etc.
The “bloomer” name is a bit silly… but, regardless, he’s on to something. We’ve been through these massive shifts (like the industrial revolution) before. What’s key is to reduce the amount of pain on the part of everyday people during the difficult transition periods that are baked in to these periods of tech advancements.
Elsewhere, Yuval Noah Harari shares more of a gloomer position (bordering on doomer, at times) suggesting that we are really getting ahead of ourselves by releasing a non-carbon-based entity into the world is all too likely to start making lots of decisions on behalf of carbon-based-lifeforms (such as ourselves). On a positive note, Harari notes instances where the U.S. has dodged the outcome of the Weimar Republic… that is, we did not fall into full-blown fascism.
My fear (based on reading many takes on this issue) is that, increasingly, there’s concern of societal breakdown during a period of mass unemployment, economic devastation, political turmoil, and the list goes on— that can result in anarchy.
Next, see my nod to a conversation with John Gray. In the conversation says, “What can be worse than tyranny? Anarchy.”
+
‘SAPIENS author Yuval Noah Harari - Humanity in the Age of AI’ (How To Academy Podcast)
Important reminder of how valuable curation has been historically… and much, much more.
For example, a realistic theory for how an AI agent given personhood could easily find a way to become the richest citizen and then lobby for more AI rights. This could lead to an AI ruling class. Not the Terminator vision of a robot takeover but a takeover, nonetheless.
*
A Guest Post by Carla Sarrett – Big Machines Don’t Cry
*
Selections of my pieces on AI over recent years…
Here’s me in June 2021 pushing back on AI Threat:
+
When Will My Job Be Replaced By AI (2023)
Spoiler Alert: Most of the jobs I was speaking of have been essentially eliminated. It took between 6-18 months for them to almost completely disappear.
+
Can Humans Determine the Quality of AI Poetry Without Bias?
+
+
The Shape of Slop to Come: Will AI Surpass Human Artistic Capabilities?
John Oliver just addressed this, in part, on Last Week Tonight.
+
It’s AI-generated, but do you like it?
+
AI Limitations & The Illusion of Explanatory Depth
*
The ground is shifting under us.
The fabric of our society is in peril.
There is still hope.
*
Tell me your thoughts, concerns, worries, fears, hopes (!?)—
I’m more than happy to do my best to answer questions about the state of AI, its progression, and where it might be headed.
Nathaniel keeps me updated on the latest from Yuval Noah Harari. It is taking so much energy just to figure out what to take in . . .as I type this N has just popped in to update!
Not long ago, ChatGPT's systems for allowing the user to use voice to interact with it got updated. I've had "conversations" with it. It's verbal output now is remarkably human-sounding, along with little chuckles, friendly banter, verbal hesitations, "Uh"s, and so on. It has made some errors but no more than I would expect from humans. There's nothing remotely "robotic" or artificial about it.
Besides all the aspects of AI I don't even know about to be concerned about, I'm most worried about (1) the rapid replacement of human work, (2) the likely total upending of K-12 and higher education, and (3) most of all the utter lack of collective preparedness and response. That last is critical. It may turn out to be the worst example ever of how a lack of international cooperation can have devastating effects. The US doesn't have its proverbial act together and shows no sign of getting it together. That, of course, makes international cooperation even more unlikely. I'm having a hard time having room for optimism.