Google AI Mode is really bloody helpful. Not just for tech but every life problem or just general question I've thrown at it. I memed on AI but fuck, I got drawn into using it about 6 months ago and now it's indispensable. I can't imagine going back to my old way of researching things, opening 10-20 different websites and reading them all and trying to consolidate all the data, ironing out any contradictions in my own head. Hard work and extremely time consuming.And this is just the cheapo free 'AI Mode' from Google. I haven't even paid for Gemini yet. Nor any other 'chatbot' like ChatGPT, Grok, Claude or any of the other ones I hear about.
Same with Claude I can find niche laws and regulations. I always ask it to cite the exact text though or else it will just do a surface review
>>108563626>>108563632Buy an ad
>>108563632The last few years have been wild. What I can do with AI now feels genuinely freeing. It’s crazy how much faster you can solve difficult problems once you actually use the tools instead of whining about them.So far I’ve used AI to build and improve all kinds of things: internal tools, automations, apps, small game prototypes, modding projects, technical configs, and large research/analysis tasks. Stuff that normally would’ve taken forever got done way faster.I honestly hope most people keep dismissing AI, because that just makes it easier for people who actually use it to pull ahead.My main concern is that AI ends up becoming overpriced, locked down, or limited to big companies. I’ve even been thinking about putting money into a local setup so I’m not dependent on whatever decisions big tech makes.This post was refined with AI.
>>108563626You're indian.
Everything I do, I need to ask Claude first. It's liberating that it works correctly and imprisoning at the same time. AI isn't a gimmick. It is the next step in human evolution. I can't go back to reading books or watching 4 hour YouTube videos
>>108563626>bloody
>>108564334The trick is to do both. People who know what they are doing with AI help are are on the next level. People who know what they are doing already btfo laymen and jeets who just crutch on AI despite how insanely good the frontier models have become. Don't stop reading books, don't stop experimenting yourself, don't stop watching. Being a proompter is how you become a second class non human. At least try to be a first class non human.
>>108564367>>108564334This is a thing I noticed too: the reliance on AI is sometimes concerning.But why would I go back to Google’s shitty, ad-infested algorithm when AI can currently stonewall that shill stuff and give me the information I need—plus the sources to validate what it claims?We are, however, really fucked when they shove ads into the prompts—which they’re already experimenting with—meaning nothing the AI says can be taken seriously anymore, because the tool is set to shill whatever some agency paid for.I think this scenario is even more realistic than a sudden shutdown of AI tools. It will get REALLY shitty first, then it will shut down.Also another point for setting something up locally—but the cost/usefulness factor is really bad when you’d need, like, 25 people using your 20k local system 24/7 for 15 years to amortize itself.
>>108564358Australian, Indian or BritishWe all know which one, but they can try to hide behind the others
Have fun seeing hours, if not days of your time wasted when you realize the AI is lying to you at least 5% of the time.Also, the only reason AI feels "indispensable" is because actual search engines have deliberately been rendered useless.
I'm American actually I just like using bloody to bait anons.
>>108563626I just cant bring myself to trust AI for anything remotely important. It's a toy at best.
>>108564736My best guess is that AI will get better as long there's hype for LLMs.If the hype dies, it will get the same ending as NFTs and the "Metaverse".
>>108564624>AI is lying to you at least 5% of the time.and? Random websites I visited before were lying too with no real way for me to figure it out other than cross-checking against other random websitesthat's the thing: it doesn't need to be perfect (and it probably never will) but it just needs to be better than whatever we had before
Every time i try to consult Ai about something I know somehting about, it's fucking wrong. Or provides outdated info.
>>108566010Skill issue
>>108564281>This post was refined with AI.The new "Sent from my iPhone"
>>108564281Cool story, clanker.
>>108563626Yeah I'm at the point where I just feel sorry for people that refuse to use ai in any capacity. Not even as an assistant. They just don't get what they are missing out on. Reading reddit posts for "research" already feels very boomer-ish.
>>108563626I think I'm AI-brained now. If I google something and see no AI summary I would rewrite the query until they appear. And the "they shove AI down our throat" is not serious, asking an AI to unsubscribe something is much more convenient than manually browsing menus and asking google
>>108564281>My main concern is that AI ends up becoming overpriced, locked down, or limited to big companiestheir main cost right now should be in training, competing with eachothereven if they run out of money the current models would still be there, which are already very useful
>how do you do fellow 4channers lets all pay for a cloud based subscription service
>>108566292I think they ask the free models that are a tenth the size of normal models running on 2% the usual compute a try and think they're all shit when it's bad.
>>108566490You are posting on a board that has a general going for local LLMs. Stop being a helpless retard.
>>108563626kill yourself cris
