AI slow down, end of woke, and China harnessing the power of the sun
Interesting trends in AI, politics, XR, and energy
Over the holidays my family all caught back to back to back flu and colds, I’m only now feeling mostly healthy again, so back to writing :)
Future of AI
As someone closely following AI advancements, there's a growing sentiment that we might be hitting a plateau. I’ve heard this in my conversations with other developers and entrepreneurs and I’ve read it all over the internet from people following the business side of AI more closely than I am.
While AI showed tremendous progress initially, recent improvements seem to be leveling out. The latest versions of AI aren’t making as significant of jumps as previous versions. The other night, I attended an AI entrepreneur startup meetup and spoke with a founder who is shutting down their AI startup. Obviously, there are a million reasons a startup can shut down, but it just kind of further highlights the trend here.
The question now on everyone’s mind is: do we need another major breakthrough technologically to reach the next big step-change in AI capabilities? Some new breakthrough beyond just feeding more data into ChatGPTs training algorithm? The current trajectory tapering off in improvements suggests that we might need a new breakthrough to reach the next level of AI capabilities. It’s possible that the current language model used to make ChatGPT will only ever be a little better than it is now. If so, we will see a cooling off in investment, with some early AI startups struggling to survive without the additional injections of capital they were depending on. It’s very possible we're entering the 'trough of disillusionment' phase of the Gartner Hype Cycle for AI, only time will tell.
My two cents? As a daily user of AIs, I feel like some marginal improvements may be all we need for AI to become the biggest disruption since the invention of the computer and software itself. As for AGI or sentient AIs from science fiction, that may require another breakthrough that maybe, just maybe we shouldn’t pursue.
Woke is in decline
I think it’s pretty safe to say that the “woke” movement peaked sometime in the recent past, with Donald Trump’s reelection serving as the key evidence for the decline, and the cherry on top of the failure of the progressive movement recently.
Post-election there’s been a lot of analysis by pretty smart people on how the progressive movement took hold of the Democratic party and eventually lost out with the electorate largely not wanting what the progressives are selling.
Interestingly, in many different studies very few Americans are progressives/woke/hard-left or whatever you want to call this group, generally it’s around 5-10% of the population. I think what’s largely been true though is that classic liberals and center-left people were having a hard time drawing a line between themselves and the progressive movement. Thus the progressive movement by default of being louder and more passionate were driving the liberal agenda in the US.
It’s purely anecdotal, but many of my previously progressive-leaning liberal friends and family have started drawing a line between themselves and the more progressive folks. Thus it may be politically “safe” now to say what many liberals have been thinking but not saying out loud, that they’re not fully on-board with all of the progressive movement’s ideological stances.
This shift is mirrored in broader trends like companies rolling back DEI initiatives and Meta ending content moderation.
This is a fascinating development in the modern politcal landscape that makes me think we’re in for some serious restructuring of the American political left.
All that being said, we’re still living in highly polarized times and neither the progressives nor MAGA are addressing effectively many issues that will have a large impact on our lives in the coming decades like the rise of AI or globalization, but more on that in future posts.
3 fun things
New cities in the US?
In a fascinating, easy-to-read article series by one of my favorite writers here on Substack, Tomas Pueyo, he explores where the US could create brand new cities based on historical analysis of why cities are successful in the first place.
35% of all game developers are making XR content according to a GDC study
It seems unbelievable, but if even partially true then XR is still on the rise, but just a slow growth.
Dropbox link of the report
China sets nuclear fusion record of just under 17 minutes
Not that long ago we were all pretty excited about a few seconds of nuclear fusion….
so nuclear fusion is progressing rapidly. If we make the final breakthroughs required for it to work it’s a game changer to us beyond what AI could ever do. Imagine limitless, nearly free, carbon-free energy?
—
if you enjoyed this newsletter and you’re not already, consider subscribing!
I write newsletters like this, occasional sci-fi short fiction, some longer essays on meatier topics, and some random off-topic stuff, but always fun, educational or to make you think in new ways about the trends happening now and in the next decade.
Wish we could sit across a table, heaped with tasty tapas treats and a bottle of Rioja and discuss the meaning of life.