The US tech industry will still be in a recession. The war against Chinese offshoring will work, but at the cost of thousands of jobs in the American tech labor industry. American businesses will become more frugal as a result and invest heavily in lobbying and future defense against the DOJ. Software startups will become more popular and less successful, hardware startups will begin waning as investors lose appetite for high-margin, high-risk entrepreneurship. ZIRP will make a return as the US realizes this, but investors will fuel a liquidity crisis that dissolves most struggling businesses.
2027 will be a more concentrated tech market with fewer small players and more money going toward "sure thing" investments in businesses with a clean track record that are too big to fail. This would be great for established high-risk brands like Tesla, provide plausible deniability for economists that only wanted to put America first, and likely crush most chances for small businesses to compete. Assuming the Trump administration intentionally gimps the DOJ (I mean, why wouldn't they?), America will be at the whims of it's own wily market.
1. There will be at least one AI personality that will become household famous.
Does Alexa count? Because that's already true then.
My prediction: Between AI search results, the various LLM platforms, and improved phone assistants, traffic to web sites drops at least 10% across the board by the end of 2025.
Hopefully not too soon. I just bought $30,000 in BITI, an ETF that tracks the inverse of Bitcoin. Basically the only way to short Bitcoin from a typical retail investment account.
I hope you're right but bubbles tend to go through an exponential blow-off phase before the crash which often takes out shorts who were right but too early.
This seems like a bad idea all around. BITI, like many “ETF that tracks ___”, tracks daily changes, which means as price chops around you’re continually losing value, i.e. BTC can break even over a year and BITI will lose value. Then there’s the issue of it tracking a deflationary instrument that is essentially guaranteed to rise over time unless it were completely eliminated, just due to people losing BTC via bad/lost hard drives.
The next administration is dealing with a high inflation environment but their policies are also inflationary adding more fuel to the fire, while the DoGE stuff is pretty unrealistic. You will probably get tax cuts funded by QE.
Inflation is at ~2%. The next admin has been handed a golden booming economy with low inflation, high wages, low jobless rate. They just don't have to fuck it up. This is a pretty low bar, but congress being congress love to doll out billions in their bills.
And I'm not saying Democrats are any better than Republicans. We are in a duopoly with an illusion of choice. Both sides love to lavish their corporate donors.
In 2024, 12-month inflation starts at 3.1% and is 2.7% last month (peak at 3.5% and bottom at 2.4%). You can't round that number in anyway to be "~2%". It is at least "~3%"
Democrats are 1000 times better than Republicans. The "both sides" rhetoric is literally propaganda.
They are not perfect by any means, but they have managed to fix every single Republican fuckup, are way more effective in passing legislation, especially bipartisan, and also dont have a record of trying to overthrow the government with a coupe, amongst many other things.
Im honestly hoping for this administration to do all the stuff they said they want to do, like tarifs, ban on abortion, removal of vaccinces, gutting of healthcare, and so on so that people who believe that both sides are bad actually feel some real pain, because thats the only way to make them understand political reality.
1. Rust lang will continue its growth, but it will will convert from curve/exponential to linear growth
2. Zig and go will continue gradual stepwise growth, due to incr adoption in the domains they already incr grow in.
3. AI in gaming hw/sw will continue to grow, with tech like onboard dlss not only being heavily marketed, but successfully executed to so as to launch certain benchmarks thru the roof, creating more shortages on top tier chips with roof busting pricepoints
4. TikTok will not be banned in the US
5. US Blue state govt climate coalitions will continue to execute despite reds in the feds. At least one trad red state will onboard into said coalition due to clear economic incentives
6. US rents will continue to increase as total % as cost of living
7. Trader Joes will sell less sweets and double fresh grocery offerings
Rust is not going to grow. It will be used for some very niche things, like cloud hypervisors where performance and security is paramount, but its just too slow to be adapted to general use case.
My prediction is that you are gonna see more ML powered transpilers/compliers that are going to be able to make high level code aot or jit compile to more optimized native code.
I think it's possible a very affordable and capable VR headset will be released (deckard?). Something with the technology of the vision pro, but competitiveness of the steam deck when it was released.
It would be a huge engineering challenge to make something like that but I think it would really rock the market.
Demand for machine learning engineers at most companies drops by 2026. I speculate that the demand for in-house models drops as the gulf between them and the major players grows. I think it will be (is?) easier and cheaper to rent time on someone else's model.
The ancillary tasks those MLEs were doing (data prep, model perf analysis, etc) will fall on other nearby roles (backend/data eng or data annalists). Ultimately I speculate that all of this work will end up being another integration backend/generalist engineers will handle.
1. There will be at least one AI personality that will become household famous.
The US tech industry will still be in a recession. The war against Chinese offshoring will work, but at the cost of thousands of jobs in the American tech labor industry. American businesses will become more frugal as a result and invest heavily in lobbying and future defense against the DOJ. Software startups will become more popular and less successful, hardware startups will begin waning as investors lose appetite for high-margin, high-risk entrepreneurship. ZIRP will make a return as the US realizes this, but investors will fuel a liquidity crisis that dissolves most struggling businesses.
2027 will be a more concentrated tech market with fewer small players and more money going toward "sure thing" investments in businesses with a clean track record that are too big to fail. This would be great for established high-risk brands like Tesla, provide plausible deniability for economists that only wanted to put America first, and likely crush most chances for small businesses to compete. Assuming the Trump administration intentionally gimps the DOJ (I mean, why wouldn't they?), America will be at the whims of it's own wily market.
A 2028 Predictions thread on HN
The sun will disappear on August 12, 2026 over Western Europe, unless the locals agree to give me treasure
Does Alexa count? Because that's already true then.
My prediction: Between AI search results, the various LLM platforms, and improved phone assistants, traffic to web sites drops at least 10% across the board by the end of 2025.
I don't think Alexa counts because I for one cannot identify:
- Any personality traits - Any apperance
Alexa has a personality trait,
it's as dumb as a box of rocks
Here’s what I found on the web for Alexa is dumb as a box of rocks.
See it does have a recognizable personality!
1. There will be at least one AI droid that is set off upon us to destroy the wealthy.
AI investment bottoms out, and it turns out that running a small 512 MB model on your phone is enough for most people.
The moat keeps drying up every time its flooded, and people are quite happy splashing around in blowup paddling pools
Programming becomes irrelevant as a career
+1, this will happen.
<table flip> Just as I am about to improve myself.
That actually happened this year. We just haven't realized it yet.
I actually agree. I think its over, but the world will take a while to catch up
Bitcoin will dramatically fall to $250k from the heights of $600k
Hopefully not too soon. I just bought $30,000 in BITI, an ETF that tracks the inverse of Bitcoin. Basically the only way to short Bitcoin from a typical retail investment account.
Any particular signal for your trade?
I hope you're right but bubbles tend to go through an exponential blow-off phase before the crash which often takes out shorts who were right but too early.
This seems like a bad idea all around. BITI, like many “ETF that tracks ___”, tracks daily changes, which means as price chops around you’re continually losing value, i.e. BTC can break even over a year and BITI will lose value. Then there’s the issue of it tracking a deflationary instrument that is essentially guaranteed to rise over time unless it were completely eliminated, just due to people losing BTC via bad/lost hard drives.
Breakthrough in prime factorization before end of 2026
Apple vision sales surpass iPad
Only if they dramatically change to a glasses based design like Meta. Otherwise I'd replace "iPad" with "Google Explorer glasses".
The Vera Rubin Observatory (https://rubinobservatory.org/about) finds observational evidence for Planet 9/X.
Very high inflation in the US.
The next administration is dealing with a high inflation environment but their policies are also inflationary adding more fuel to the fire, while the DoGE stuff is pretty unrealistic. You will probably get tax cuts funded by QE.
Inflation is at ~2%. The next admin has been handed a golden booming economy with low inflation, high wages, low jobless rate. They just don't have to fuck it up. This is a pretty low bar, but congress being congress love to doll out billions in their bills.
And I'm not saying Democrats are any better than Republicans. We are in a duopoly with an illusion of choice. Both sides love to lavish their corporate donors.
In 2024, 12-month inflation starts at 3.1% and is 2.7% last month (peak at 3.5% and bottom at 2.4%). You can't round that number in anyway to be "~2%". It is at least "~3%"
being 100 bps off target is not a "high inflation environment"
My Parmesan cheese is now $4, up from $3.50, price changed the day after thanksgiving.
I don't think inflation is low right now. They're working overtime to make it look low.
Definitely more 3% than 2.
Ill say it since it people seem to be confused.
Democrats are 1000 times better than Republicans. The "both sides" rhetoric is literally propaganda.
They are not perfect by any means, but they have managed to fix every single Republican fuckup, are way more effective in passing legislation, especially bipartisan, and also dont have a record of trying to overthrow the government with a coupe, amongst many other things.
Im honestly hoping for this administration to do all the stuff they said they want to do, like tarifs, ban on abortion, removal of vaccinces, gutting of healthcare, and so on so that people who believe that both sides are bad actually feel some real pain, because thats the only way to make them understand political reality.
Unemployment rate goes all time high.
Trump either defuses the war(s) in the middle east or it accelerates and becomes nuclear.
Dedollarization accelerates
1. Rust lang will continue its growth, but it will will convert from curve/exponential to linear growth 2. Zig and go will continue gradual stepwise growth, due to incr adoption in the domains they already incr grow in. 3. AI in gaming hw/sw will continue to grow, with tech like onboard dlss not only being heavily marketed, but successfully executed to so as to launch certain benchmarks thru the roof, creating more shortages on top tier chips with roof busting pricepoints 4. TikTok will not be banned in the US 5. US Blue state govt climate coalitions will continue to execute despite reds in the feds. At least one trad red state will onboard into said coalition due to clear economic incentives 6. US rents will continue to increase as total % as cost of living 7. Trader Joes will sell less sweets and double fresh grocery offerings
Rust is not going to grow. It will be used for some very niche things, like cloud hypervisors where performance and security is paramount, but its just too slow to be adapted to general use case.
My prediction is that you are gonna see more ML powered transpilers/compliers that are going to be able to make high level code aot or jit compile to more optimized native code.
Slow in terms of performance or development velocity?
I think it's possible a very affordable and capable VR headset will be released (deckard?). Something with the technology of the vision pro, but competitiveness of the steam deck when it was released. It would be a huge engineering challenge to make something like that but I think it would really rock the market.
The Meta Quest is already incredibly affordable. If that can't make VR mainstream, I'm not sure what can.
Robo helper in the household
Ai personal assistant who knows me and understands what I want
Ais chat among each other to figure out if their “owners” are good matches. If they are the ais set up a time and place for the first date
So… Black Mirror season 3 then?
Ali Khamenei dies
Netcraft will confirm Reddit is dead.
Demand for machine learning engineers at most companies drops by 2026. I speculate that the demand for in-house models drops as the gulf between them and the major players grows. I think it will be (is?) easier and cheaper to rent time on someone else's model.
The ancillary tasks those MLEs were doing (data prep, model perf analysis, etc) will fall on other nearby roles (backend/data eng or data annalists). Ultimately I speculate that all of this work will end up being another integration backend/generalist engineers will handle.