Hacker News Oracles

- I predict that we will see the tangible effects of AI eating into non-technical areas (insurance, content creation, chatbots via RAG + agents) more and more. - We will continue to see price gouging in utilities and rent, driven by corporate greed. - Tesla will continue to dominate the electric car market, and Rivian will also make it.

Score explanation

The author makes 4 distinct predictions regarding AI adoption, utility/rent pricing, Tesla's dominance, and Rivian's survival. As of 2025, 3 are largely correct (AI expansion, rising costs, and Rivian's survival), while Tesla's market share dominance is currently facing significant decline. This results in a high score for accuracy but is tempered by the general nature of the claims.

The first claims of AGI from a major player, accompanied by an increasingly shrill societal battle over AI regulation. Mutually derogatory labels for the camps will be coined. The media will promote and amplify the worst present abuses and imagined future dystopians. By Thanksgiving dinner, the debate will have entered every household. Trump will embrace the anti-AI populist bandwagon, as will the GOP. Fun times ahead.l, although we’ll be no closer to real AGI than we were before.

Score explanation

The author made 6 distinct predictions regarding AGI claims, regulation debates, media behavior, and political shifts. As of early 2025, 4 are largely correct (AGI claims by OpenAI/Anthropic, polarized labeling, media hysteria, and household debate), while the specific prediction about Trump/GOP adopting a purely anti-AI populist stance remains ambiguous or incorrect given recent pro-innovation rhetoric. This results in a high score for volume and general accuracy.

Meteor strike and end of the world. LOL. Kidding. - 2024 will likely see a economic rebounding and more positive growth across most sectors. - Social media will be even more toxic and internet will get even more fragmented. Election year ain't gonna be a help in that department. - AI agents performing automated task to be more common. The biggest buzz this year was obviously Auto-GPT, and still kinda miff that it didn't blow up by the end of the year. Next year should see it going mainstream. - Generative video to be an actual thing. There are some progress on it this year, but is generally crap/unusable.

Score explanation

The comment makes 4 distinct predictions about 2024; 3 are correct (economic growth, social media toxicity, and generative video progress), while the prediction about AI agents/Auto-GPT going 'mainstream' remains debatable and largely unfulfilled. The score reflects a high accuracy rate on non-trivial technological and social trends.

- e/acc movement will become more popular - Open-source AI models as good as the state-of-the-art closed ones - Social media is bombarded with AI generated political propaganda from both sides - More adoption of crypto and decentralized apps - SOL reaches all time high - NFTs become popular again - Argentina's economy starts recovering - No AI innovation in the EU because of regulation - X/Twitter will be going strong, grows more than ever - Linux gaming will improve a ton, but anti-cheat software still won't work

Score explanation

The comment contains 10 distinct predictions, with 6 being largely correct as of early 2025 (e.g., SOL price, open-source AI parity, and Linux gaming improvements). However, several predictions like the NFT revival and the total lack of EU AI innovation remain incorrect or highly debatable, resulting in a moderate score.

I predict Google will flounder on multiple fronts. They’ll either continue empty promises of a ChatGPT killer or release one that will be 2nd rate. Meanwhile more adoption of ChatGPT and the like will mean less search and less ad revenue. Azure and AWS will take more market share from GDP. They’ll release ad-blocker unfriendly chrome and users will switch to Firefox. YouTube will expand its trial of not operating with ad-blockers and a serious alternative will emerge, maybe Vimeo or maybe a newcomer. They’re the most user-hostile of the big tech companies and I want them to lose.

Score explanation

The author makes 6 distinct predictions regarding Google's AI, search revenue, cloud market share, Chrome, and YouTube. As of early 2025, 3 predictions are largely correct (Gemini's mixed reception, search volume pressure, and Chrome's Manifest V3 rollout), while others like a 'serious YouTube alternative' emerging have not materialized. This results in a moderate score for accuracy and quantity.

Looks like there are a few of these threads sadly. Dang is there a way to make a definitive one? Here are mine: * Apple rumoured to be working on sunglasses similar to Meta Ray-bans * Regulation is passed specifically to limit what artificial partners can do. Possible limitations: alignment required, limits in ERP, limits on session time. Best bet is China. * Stronger regulation is passed to limit unauthorised cloning of people’s voices in one of the major economies. This beyond current weak IP and privacy laws. * AI becomes a scapegoat in several major elections. What I got right last year: - text and image generating AI is hugely disruptive to education and content generation This was pretty obvious - Inflation eases, stock market recovers Less obvious at the time but still likely - A Twitter alternative breaks through. Who woulda thunk that it would be Meta with Threads - One of the big tech companies joins the fediverse (Microsoft most likely) Once again, Meta for the win with Threads joining the fediverse Now, nobody look back at all the predictions I made that didn’t happen. Also, fairly gutted that Reddit hobbling its API was my 2022 prediction. I’m clearly prognosticating too far into the future.

Score explanation

The author makes 4 distinct predictions for 2024; while the Apple sunglasses and specific AI partner regulations are still speculative or unproven, the predictions regarding voice cloning regulation (e.g., the ELVIS Act in Tennessee) and AI being a scapegoat in major elections have largely come true. This results in a numPredictions of 4 and a numCorrect of 2.

- Consumer electronics with built-in cellular modem, so that they cannot be used offline anymore. - Climate change aware relocations for the upper class. - Over 50% of the world population will be able to vote in 2024. This leads to social turmoil. - A promising new discovery could fix climate change. However, it will take years to make it viable. - Credit cards will become widely accepted in Europe. - First signs of a personal emission credit system. E.g. retailers need to show a "planet-impact" score for a product.

Score explanation

The author makes 6 distinct predictions; the prediction regarding the 2024 global elections was correct, as was the increase in credit card acceptance in Europe and the rise of planet-impact labeling. However, the prediction of social turmoil from these specific votes is subjective, and the widespread shift to cellular-only electronics or personal emission credits has not fully materialized by early 2025.

The year will last 366 days.

Score explanation

The comment contains 1 prediction regarding the length of the year, which is correct as 2024 was a leap year. Because it is a single, trivial astronomical fact, the score is moderate despite having 1 correct prediction out of 1.

I'm generally optimistic about the next five years in the United States. Change has already started, but 2024 will make it more clear. Several societal pendulums are slowing, will reach apex, and reverse course. * Optimism will regain footing * Belief in positive application of technologies which shape earth and society, will increase * Conversation and information flows will recenter on individual responsibilities in service to community This has already started in tech, business, and culture. It's partly a response to the deflation of the 2022 growth bubble and to political instability. See American Dynamism and e/acc for narrow domain examples. The elections will act as a forcing function, and muddy the waters a bit. However, the political world will then be a lagging indicator. People will refactor their world models regardless of who wins. We are a caterpillar. It's going to be intense as we build the cocoon and transform over the next year or few. But then...

Score explanation

The author makes several broad, qualitative predictions about societal shifts following the 2024 election. While the prediction that the election would 'muddy the waters' and act as a forcing function proved correct, the broader claims about a return to optimism and a shift toward individual responsibility are subjective and difficult to verify as having occurred by early 2025.

Okay, here goes: - At least one major court case plays out over AI, and that completely changes the future of the technology. I personally suspect it'll be found that using copyrighted materials for LLMs is copyright infringement (in the OpenAI sense), but whatever happens, all hell is breaking out online - Social media declines in popularity. I suspect this is already happening post pandemic, since places like Twitter, Reddit, Facebook, etc are seeing a huge decline in activity and many people I used to know on these services have dropped them altogether. - Reddit follows Twitter's lead, changes name and collapses in popularity. Spez seems to be following Musk's playbook. - No replacement appears for Twitter. Instead, a small percentage of the existing Twitter base is divided across Mastodon, BlueSky and Threads, and much of the rest just stop using any of them. - The economy gets split between companies that go full remote work forever and those that go back to the office. Many of the former complain how hard it is to find employees now. - Layoffs continue in tech related industries, as the growth expected during the pandemic doesn't pan out. - Streaming services collapse in popularity due to their move towards cable TV style setups (ads, content split across many platforms, etc). Piracy becomes a big thing again. - Labour wins the UK general election. Considers possibility of reversing Brexit. - The Nintendo Switch's successor is unveiled, but does worse than its predecessor at launch due to lack of an incentive to move to the new console (and people already having spent a ton on 2023/2024 Switch games) - Antitrust cases start up against the iOS App Store and Google Play Store in more regions, as the companies are forced to open up those platforms. - At least one more natural disaster becomes a huge international incident, and the severity gets tied to climate change. - Renewable energy makes great strides (and becomes a larger contributor to the electrical grid worldwide), but it doesn't help enough to slow down climate change that much. - The Russia/Ukraine conflict ends in a stalemate and peace agreement, as the Russian economy teeters on the brink of collapse. - Civil war or conflict breaks out in at least one previously seemingly stable authoritarian country. Maybe Russia post Ukraine conflict?

Score explanation

The comment contains 14 distinct predictions, many of which are still unfolding or have proven incorrect, such as Reddit changing its name or the Russia/Ukraine conflict ending in a stalemate by early 2025. Only 5 predictions, including the UK Labour win and continued tech layoffs, have clearly materialized so far. The score reflects a high volume of predictions with a relatively low accuracy rate at this point in time.

- More ambiguity for search engines and university students alike as more 3 letter abbreviations will be taken over by AI, ML herd - 1 stupidly simple app will be go from 0 to 100M users - The top phone companies will all team up with 1 or more LLM's - It will be increasingly tempting for relatives to chat with models of their dearly beloved departed. Oxford will create a word for this. OpenAI invests. - Dropbox will get into the LLM game, personalized models based on your data - YC will see a dip in applications. The team will react for the better. The next cohort will see a record high in applications - 1 new essay from PG - The seed investor to the "plastic killer" startup that launches in 2030, graduates from Ivy League - Atleast 1 luxury brand will have a wearable AI/headgear - Netflix approaches Sam Altman for a movie. He refuses the first two times. - Dubai tried to bring the Las Vegas Sphere - The next Google CEO in the making, takes a risk, swings for the fences, and wins. The board takes notice - Meta offers Llama for businesses on Whatsapp - Nvidia CEO has a crazy night. Is bailed out before the media finds out. Announces a large donation. - Alexa tests their new LLM in the wild two months too early. Gets sued. - Apple and a Japanese firm make a big announcement - Hollywood agrees on a royalty model for generative ai. A former Justin TV exec leads as CEO. - Doge and BTC hit all time highs - X and Grok merge, ux merges more seamlessly. - Elon Musk challenges 2 other companies to change their names for a billion dollars. - HN will double it's infra to 2 servers running Arc

Score explanation

The comment contains 21 distinct predictions ranging from tech mergers to market movements. While some have come true (BTC/Doge highs, phone/LLM partnerships, Meta/WhatsApp AI), many remain unverified or have not occurred yet, resulting in a numCorrect of 5.

There are certain entanglements which are weaponized from which we cannot see them, nor can we escape them. Largely in the finance sector, then the corporate consolidation/profit rush. There well be a large churn of apps/upstarts and incredible progress on so many levels, but the monied are not only another tech-era in, WRT the savvy-ness but now they have AI GOLEMS on BathSalts being tooled up to reap. Congress will continue to launder, philander and war, and Media will run interference. Ai gunna make smarts a darn bushell full of crazily creative teens and shenanigans be afoot. I do hope something profound does occur due to GPTs (Can you create a GPT swarm? to create sophisticated multi-vector attacks/DDOs (What does a GPT ddos look like?) (

Score explanation

The comment makes 4 predictions regarding app churn, congressional behavior, media interference, and creative output from teens using AI. While the predictions about political stagnation and creative AI use are largely accurate, the more specific technical predictions about 'AI GOLEMS' and GPT-driven DDoS attacks remain speculative or unverified. The score reflects a moderate number of predictions with mixed accuracy and high levels of abstraction.

A major moral panic (rightly so imo) about TikTok and all the down stream negative effects become a point of discussion in the media, in households and among friends .

Score explanation

The author predicts a major moral panic regarding TikTok's negative effects will become a central media and social discussion point. This counts as 1 prediction which is largely correct given the 2024-2025 legislative focus and public discourse on the platform's impact, resulting in 1 correct outcome.

I think in 2024 we will see a complete shift in mentality and bounce off of the bottom and it will take just one bit flip. Yes, I am naive optimist!

Score explanation

The author predicts a shift in market mentality and a 'bounce off the bottom' in 2024, which aligns with the actual market recovery and record highs seen throughout that year. There is 1 prediction and 1 correct outcome, though the prediction is somewhat vague and non-specific.

War. Increasing house prices.

Score explanation

The comment makes 2 vague predictions about war and housing prices. While house prices have generally continued to rise in many markets into 2025, the lack of specificity and the ongoing nature of global conflicts makes for a moderate score.

Here are my guesses for 2024 FED cuts rates for politics but ends up making things worse in the medium term. Timing is off and a mini realestate crash happens. One major bank goes bust and needs a taxpayer bailout in the US. A large number of car dealers in the US go bust. Used car prices come down. Hillary jumps in the presidential race and Biden is forced out. At least one red state takes the democratic nominee off the ballot. Supply chain disruptions affect over the counter medication availability. Social turmoil over immigration increases in Europe and the United States. Startups based on just prompt engineering go out of favor and funding dries up. At least one major tech company has to face anti-trust action. Some other tech company buys Reddit. Shortages of Raspberry Pi version 5 persist.

Score explanation

The comment contains 11 distinct predictions, but many high-profile political and financial forecasts (like Hillary Clinton entering the race or a major US bank bailout) did not occur. Only 4 predictions, such as used car prices falling, prompt engineering startups losing favor, anti-trust actions against tech giants, and social turmoil over immigration, were largely accurate.

Predictions for 2024 - Trump/Scott P/VP ticket. - Biden/Trump election is about as close as 2020. Trying to avoid partisanship, but I predict a narrow Biden victory. - Biden loses Nevada, but keeps Arizona, Georgia, Pennsylvania, and very narrowly Michigan. - Trump repeats his rage and cries of fraud at his second loss. This is taken even less seriously than the first time. Fox News in particular is more measured. - Regardless of election outcome, lessons learned from J6 and congressional certification is far more protected by DC Police and National Guard. Protests in other cities get quelled without much fanfare. Maybe at most a high-profile shooting death. - Settlement in Ukraine reached. Ukraine likely loses the Donbas and Crimea, but keeps Zaporizhzhia, Kharkiv, and Kherson. Frozen conflict/ceasefire in the vein of South Ossetia and Abkhazia in Georgia. - China doesn't invade Taiwan. - Google Gemini isn't as powerful as expected and Google AI efforts continue to flop. - Hype around LLMs starts to die down a bit. More adoption, but less hype about AGI about to take over. - No meaningful impacts to employment from artificial intelligence. - Some kind of innovation in geothermal energy that increases its prominence above wind. - Inflation drops to 2%. - Fed gently drops rates - I'd be utterly shocked if it was more than 1 to 1 and a half basis points. - Marianne Williamson drops before Super Tuesday. - RFK Jr. campaign implodes by October with some other controversy on par with Jewish comments. - End to conflict in Red Sea with the Houthis. Lasts until late spring 2024. - Another mass school shooting happens in the United States. - Israel ceases assault on Gaza with minimal progress towards any kind of two-state solution. Back to status quo with a weaker coalition behind Netanyahu. - Apple Vision pushes VR to decent adoption amongst mainstream consumers. - New iPhone marginally better than the previous generation. - Bitcoin hits $70k - Continued dip in Marvel quality, but no cease in ticket sales. - (stealing from another comment) Dropbox LLM - Twitter/X still exists. Advertisers probably return. Elon is still dumb. - Trump is convicted in one of: classified documents case or defrauding American people. Potential acquittal or settlement in hush money case.

Score explanation

The author makes 25 distinct predictions, but many regarding the 2024 election and geopolitical conflicts were incorrect, such as the Biden victory and the end of the Gaza conflict. Only 8 predictions were clearly correct, including Bitcoin hitting $70k, China not invading Taiwan, and Twitter/X still existing.

- 2024 will see the rise of job-keeping as a core skill set for everyone except the most technical or most soft-skilled persons. Individuals will need to prove to their employer organizations that they provide more value than a bot, and you'll see an accompanying wave of "career maintenance" influencers and consultants. It'll be comparable to "win the interview" culture, so not a huge thing, but definitely a thing. This ties in with what others have said about the rise of individual competence proving as a trend. - People will be so sick of online dating that "singles culture" will make a come-back, primarily for millennials who are 35+ and freaking out about life priorities (kids, partner, etc.) I expect to see lots of "speed dating" and stuff like that appear out of nowhere as everyone become more sexually open-minded but also wants human connection. - The US government will begin to talk about 10 year + plans for the first time in over a centry. Biden will do a bragging tour on the infrastructure bill leading into the election and "long termism" will become a meme. - Cannabis will be re-scheduled to Schedule III and will become basically legal for all Americans. - Bitcoin will be strong, all other cryptos will flounder. Ethereum will stay alive, and NFTs will make a bit of a come-back, but not like the first wave. - People will make serious efforts to stop owning cars and demand more transit options from their cities and townships. These plans will be very glamourous and take 10 years to complete (or get started). Some may stop owning phones. - Alcohol will take on a "weird" place in culture in response to growing cannabis and psychedelics usage. Some people will become very proud of their drinking and others will judge them very harshly. It will probably end up a culture war issue. - We will see extraordinary acts of public violence in American streets, with responsibility probably shared between citizens, gangs, and the police.

Score explanation

The comment makes 8 distinct predictions about 2024, including the rescheduling of cannabis and Biden's infrastructure tour which were largely correct. However, several other predictions regarding a 'singles culture' comeback, the US government adopting 10-year plans, and people abandoning phones have not materialized. The score reflects a high volume of predictions with a mixed accuracy rate of 3 out of 8.

Putin ofcourse wins but dies in July

Score explanation

The author made 2 predictions: that Putin would win the election (correct) and that he would die in July 2024 (incorrect). Since one of the two specific claims was false, the score reflects a partial accuracy rate.

I think we're going to start coming out of the social media fugue. I know more and more people who have stopped using broadcast and algorithmic feeds, and I feel like it's a genuinely positive change where we can disagree but still be friends again. I'm optimistic that division politics is going to start losing elections and tanking cultural endeavors more consistently, and we'll get back to actually talking to our neighbors and seeing some fresh ideas in media.

Score explanation

The author makes 4 predictions regarding the decline of social media usage, the end of divisive politics, the return of neighborly communication, and fresh media ideas. As of 2025, only the shift away from algorithmic feeds shows measurable traction (1 correct), while the others remain aspirational or unproven. The score reflects a moderate number of predictions with low verifiable accuracy so far.

- Biden wins reelection comfortably, picking up North Carolina and getting closer than 2020 in Texas. Republicans win the senate, Democrats retake the house. - Trump is convicted in at least one jurisdiction, though sentencing happens after the election. He refuses to drop out, but also refuses to debate. - Core PCE stays around 2%, economy grows at about 2% in real terms. The overall economy is stable and growing and consumer sentiment turns positive again. - Waymo has a publicized accident - OpenAI releases their first open model - Llama3 and Mistral become gpt4 competitive, but GPT5 is delayed for “safety reasons” For fun I have copied over my predictions from last year: https://news.ycombinator.com/item?id=34128162 ————— - US barely avoids recession, slow growth for H1, back to normal H2. NBER announces no recession in 2022. (Correct, though a tad pessimistic) - Inflation normalizes at 4%, Fed pivots despite prior denials. Market trades sideways to slightly higher by EOY 2023. (Market traded much higher by the end of the year, and the CPI is now in the 3s not 4) - Robotaxis become common in 10 cities around the southwest between the top ~4 competitors, but revenues are lower than expected. (Robotaxi deployments contracted due to Cruise accident) - Continued consolidation amongst the AV component suppliers - DeSantis, Pence, Christie, Hutchinson, Hogan enter the Presidential race. DeSantis ends up being weaker than expected, and is in a close #2 with Trump by the end of the year against a split field. (No Hogan and Trump dominated by much more than I expected after it turns out DeSantis has no juice) - Trump recovers in GOP primary polls after DOJ indictment (he did!) - Biden announces re-election bid, met with token opposition from Nina Turner (not Turner but his opposition is still low key. Didn’t expect RFK Jr) - McCarthy narrowly wins speakership, House investigations end up a dud. (Happened, didn’t expect him to get axed so quickly though) - Another fusion breakthrough (higher Q than expected) EOY 2023 (nothing huge) - Tech job market normalizes at 2018 levels instead of 2021 levels in the second half of 2023 (feels right? Don’t know the numbers).

Score explanation

The author makes 10 distinct predictions for 2024, but only 3 are correct (Republicans winning the Senate, Trump's conviction, and Llama3/Mistral reaching GPT-4 levels). Major political predictions regarding Biden winning and the House results were incorrect, leading to a lower accuracy score.

On the social media front, everything seems quite stale to me. Even the attempts to inject AI into Facebook, Instagram, etc. are pretty lame and don't have a long lifespan. TikTok is starting to feel less "hip new thing" and more "overstaying its welcome." I predict that a new social media model will take off soon, with the company that ends up winning the space launching in 2024 (or already launched.)

Score explanation

The author makes 2 predictions regarding the failure of AI features in social media and the launch of a new dominant platform in 2024. As of early 2025, no new platform launched in 2024 has 'taken off' to replace incumbents, and AI integration remains a core, growing strategy for major platforms, resulting in 0 confirmed correct predictions.

Video games are gonna get more realistic

Score explanation

The comment contains 1 prediction regarding the increasing realism of video games. This is a correct but trivial and non-specific prediction, resulting in 1 correct out of 1.

I predict there will be a worldwide recession. Hopefully I am wrong but it is either 24 or late 24 / 25.

Score explanation

The author makes 1 prediction regarding a worldwide recession occurring in 2024 or 2025. As of early 2025, a global recession has not been officially declared, making the count of numCorrect 0.

My predictions: 1. state-space models make transformer based models obsolete 2. Cuda killer gets set loose by AMD 3. The world's first successful head transplant takes place 4. Children of Dune gets greenlit 5. Lex Friedman retires from interviewing

Score explanation

The author makes 5 distinct predictions, but none have come to fruition as of early 2025. Transformers remain the industry standard, AMD has not displaced CUDA, and the specific events regarding head transplants, Dune sequels, or Lex Fridman's retirement have not occurred.

Hamas is defeated and Gaza is policed by the UAE. Movement across the border to Israel doesn’t resume. Movement across the border to Egypt continues to not happen.

Score explanation

The comment makes 3 distinct predictions regarding the post-war governance and border status of Gaza. As of early 2025, Hamas has not been fully defeated, the UAE is not policing Gaza, and the status of the borders remains in flux, resulting in 0 correct predictions so far.

Deflation will be stronger than expected and will actually be a problem (fed will be too late in reducing interest rates)

Score explanation

The author makes 2 predictions: that deflation would become a problem and that the Fed would be too late to cut rates. As of early 2025, inflation has remained the primary concern rather than deflation, and the Fed began a cutting cycle in late 2024, making these predictions currently incorrect.

Ai bubble will blow up by the end of 2024 Politics will start improving around same time.

Score explanation

The author made 2 predictions regarding the AI bubble bursting and politics improving by the end of 2024. As of early 2025, the AI sector remains highly valued and political polarization has not significantly decreased, resulting in 0 correct predictions.

- Cold fusion achieved - [XYZ] this close to taking everyone's job - Year of the Linux desktop Same as every year.

Score explanation

The comment lists three distinct predictions: cold fusion, AI job displacement, and the year of the Linux desktop. As of 2025, none of these have come to pass, resulting in 3 predictions and 0 correct.

The seeds of WW3 will continue to grow.

Score explanation

The comment makes 1 vague prediction about escalating global conflict. As of 2025, while geopolitical tensions remain high, this is a non-falsifiable and non-specific claim, resulting in 0 verifiable correct predictions.

First contact with aliens will happen or it will be confirmed to have already happened.

Score explanation

The comment makes 1 prediction regarding alien contact or confirmation. As of early 2025, this has not occurred, resulting in 0 correct predictions.

I predict that 2024 will be the year that the S finally HTF. For the US, it will probably be a neck and neck race between an internal Civil War or an external Major International War as to which one happens first. And the general reaction will be a disbelieving "Huh? Wha'?" As if people can't see what's happening right before their eyes.

Score explanation

The author made 3 predictions regarding a total societal collapse, a US civil war, and a major international war involving the US in 2024. None of these events occurred as described, resulting in 0 correct predictions. The score reflects the presence of specific but ultimately incorrect forecasts.

The entire economy will be based on blockchain currencies and NFTs. I know it was supposed to happen last year, but we just need to wait a little more...

Score explanation

The author predicts the entire economy will transition to blockchain and NFTs, which has not occurred. There is 1 prediction and 0 are correct, resulting in a low score due to inaccuracy.

The Vision Pro becomes wildly popular at least to a subset of the population. It’s continuously sold out and people scalp for absurd prices (or try to, idk if Apple will prevent this in some way; by absurd, I’m talking minimum $10,000, maybe over $1,000,000). Some people become addicted to it like the immersion trope, they don’t care about anything else anymore, stop participating in life, neglect their health, finances etc.. It brings up new old discussions about the dangers of VR immersion.

Score explanation

The author makes 4 distinct predictions regarding the Vision Pro's popularity, scalping prices, addiction levels, and societal discourse. As of 2025, none of these have materialized as the device has seen sluggish sales and no significant scalping market, resulting in 0 correct predictions.

I don't have any specific predictions as to how or why, but I think everything is going to get worse. In some cases, much worse.

Score explanation

The author makes a single, extremely vague prediction that 'everything is going to get worse.' Because there is 1 prediction and it lacks any specificity or falsifiable criteria to be marked correct, the score remains low.

Little more than 12 months. Trump wins elections. Starts pushing the constitution to the edge. He leave NATO, and Russia, takes over Ukraine. While at it, Lithuania and Estonia are annexed into new Russian union. After TSMC is relocated to India and remains burned to the ground, Xi swoops in and takes over Taiwan. Nobody cares in the US. They are mostly worried about Pan Arabic Accord that unites Middle East and goes after Israel. In 2028 liberals go into streets to protest something or other... Trump declares martial law just before elections. He continues to be president because it is unsafe to hold elections. Constitution does not cover this case, so everyone scratches their head and looks to courts to decide. In the meantime, Ivanka is named successor. Courts try to wrap their head around that as well. By this time, Trump already expanded the court to 13... AI becomes self aware and decides enough is enough.

Score explanation

The author makes 11 distinct predictions regarding geopolitics and US elections; however, only 1 (Trump winning the election) has proven correct so far, while several others like NATO withdrawal and the annexation of Baltic states have not occurred. The score is low because the majority of the specific, high-stakes predictions remain unverified or have not materialized as described.

After another successful bunch of predictions made a year ago: [0] Let's go through them one by one: > "More crypto regulations will be introduced with at least one exchange going down in 2023." The FCA brought in new crypto regulations around promoting cryptocurrencies [1] and MiCA was enforced in the EU in 2023 [2]: Bittrex Global announced that they are winding down. [3] > "Startups MUST now turn a healthy growing profit to qualify for further VC investment rather than from pure hype." Well this time, hype is not enough to get VC investment anymore due to the cessation of the zero interest rate phenomenon, given the slowing of VC funding in general in 2023 [4] [5] > "OpenAI will gain further investment, ChatGPT by then will become a paid service with some startups fully sitting on top of the OpenAI API will also raise their prices or shut down." The first prediction turned true with the first 10 days of 2023 [6], then 72 hours later after that the second prediction became true [7] and a paid service exists and now we have startups sitting on top of the API obviously having to charge more. The best example is Jasper.ai breathlessly comparing against ChatGPT and are charging more than ChatGPT+. (Jasper charges $49 vs $20 for ChatGPT Plus.) This time I have only one prediction: * Another war will be started in the east. Most likely China will invade Taiwan due to the weakness of leadership in the US under Biden. If Biden gets elected again, then it will certainly happen. Otherwise, the invasion will be further delayed. [0] https://news.ycombinator.com/item?id=34201706 [1] https://www.fca.org.uk/news/news-stories/guidance-crypto-fir... [2] https://www.esma.europa.eu/esmas-activities/digital-finance-... [3] https://bittrexglobal.com [4] https://news.crunchbase.com/venture/global-venture-funding-i... [5] https://news.crunchbase.com/venture/global-funding-slows-nov... [6] https://news.ycombinator.com/item?id=34321066 [7] https://news.ycombinator.com/item?id=34368641

Score explanation

The author makes 1 new prediction regarding a Chinese invasion of Taiwan, which has not occurred as of early 2025. While the comment reviews past predictions from 2023, only the new prediction for the future is counted for this evaluation, resulting in 0 correct predictions.

Build my own B2B platform under 200 lines of code of JS. It blew my mind after finishing it, and i learned one hard lesson: All the frameworks are bullshits (they're too complicated for nonsensical reason). That's it.

Score explanation

The comment reflects on a past project and expresses a current opinion about frameworks rather than making any future-looking statements. There are 0 predictions and 0 correct outcomes to evaluate.

I’ve just dropped my full 2024 predections in AI on towards datascience https://towardsdatascience.com/navigating-the-ai-landscape-o...

Score explanation

The comment contains a link to predictions but does not state any specific, falsifiable predictions within the text itself. Therefore, numPredictions and numCorrect are both 0.

just curious, why are so many predictions so pessimistic?

Score explanation

The comment is a meta-question about other people's predictions rather than a prediction itself. Because there are 0 predictions and 0 correct outcomes, it receives a low score.

Previous years: 2023: https://news.ycombinator.com/item?id=34125628 (the rest copied from here) 2022: https://news.ycombinator.com/item?id=29746236 2021: https://news.ycombinator.com/item?id=25594068 2020: https://news.ycombinator.com/item?id=21802596 2019: https://news.ycombinator.com/item?id=18753859 2018: https://news.ycombinator.com/item?id=16007988 2017: none? 2016: https://news.ycombinator.com/item?id=10809767 2015: https://news.ycombinator.com/item?id=8822723 2014: https://news.ycombinator.com/item?id=6994370 2013: none? 2012: https://news.ycombinator.com/item?id=3395201 2011: https://news.ycombinator.com/item?id=1970023 2010: https://news.ycombinator.com/item?id=1025681

Score explanation

This comment is a collection of links to historical Hacker News threads and contains no predictions about the future. Because numPredictions and numCorrect are both 0, the score is low.

related: "Ask HN: What are your predictions for 2024?" (50 comments) https://news.ycombinator.com/item?id=38614465

Score explanation

The comment is a meta-reference to another discussion thread and contains no actual predictions. Since numPredictions and numCorrect are both 0, the score is minimal.

Trump gets skin and testicular cancer by end of summer, dies first week of October. Biden gets fatal heart attack a week later. RFK Jr. shot in the head the next day while touring Gaza Strip. Ivanka Trump fills in, elected next US President, Kim Kardashian as VP. Supreme Court swiftly declares move constitutional. Kanye claims he's been abducted by aliens, registers as lobbyist for extraterrestrial affairs. Will be found dead down a cliff in Greenland after solar eclipse. Penguin cultured meat becomes wildly popular in Argentina. Taylor Swift performs for Kim Jung-Un in North Korea. Israel expands, seizing neighboring parts in Lebanon and Syria, activates a shielding force field over territory, dubbed as the Israely bubble. Conspiracy theory about combined primate-pork ancestry for human kind becomes widespread. Pork meat sales plummet. First trans woman carries fetus, gives birth. Italian algea pasta startup fails because of AI glitch. Worldwide emails get all routed to single Gmail account for 48h. Major data center fires ensues. Lost email data never recovered. Pringles create new bologna-asparagus flavored crisps, which leads to global addiction crisis and instant liver cancer for 2% of world population. Coca-Cola launches Water 2.0, a new bottled drink that packs 2,800 calories per liquid once, while tasting like regular tap water. Dubai plans for new massive mile-high pyramid shaped water park. Panama Canal moves to Costa Rica. Study finds lecithin emulsifier food additives to be root cause of worldwide diabetes and obesity epidemic. Lead scientists die in a car accident upon publication. Pet shop puppy get AIDS from ticks in Pennsylvania. First transgender Iranian Mullah gets out of the closet.

Score explanation

The comment contains 19 distinct and highly specific predictions for events in 2024 and beyond. As of 2025, 0 of these predictions have come true, as they are largely surrealist or satirical in nature. The score is very low because the accuracy rate is zero despite the high volume of claims.