Funny that article is in futurism.com that is jam packed with tracking cookies and fingerprinting.
I want the personalization…without Brother Brother. My personal and local AI, should work solely for my interests. I have been using my local AI for creating an customized cookbook, making save editors, and right now, an restoration mod.
Never before have I had so much agency in my life. AI has great potential for self actualization, but corporations and government will be working to undermine it. Hopefully, everyday people will recognize that and make local AI the norm.
Exactly. The real promise of AI isn’t having a corporate assistant that remembers your life so it can monetize you. It’s having a tool you control, running on hardware you own, helping you create things you actually want. Local AI turns personalization into empowerment rather than surveillance. Which is probably why Big Tech would prefer to rent you your own brain by the month. How does your setting look like? What hardware, harness, agent, modell are you running locally?
Currently, using the Strata backend with Qwen-Flash-Next. It has successfully made 4 save game editors, 2 plugins, and did some work on La-Mulana 2006 to make the controls and graphics play nice with Windows 11, plus my gamepad. The DLC restoration mod is largely successful, I am doing QA for making the DLC accurate to what was offered in the ad copy.
000000000
Hermes Agent is my model of choice, and I have Strata using 256k context, on a IQ3. I find that Next-Flash is much faster than the smaller 27b, and has more personality. Sometimes it can get a bit loopy. I typically have the AI keep a bible of findings, and start a fresh Hermes session - then tell it to condense the information from the old workspace and session into a more relevant version. Save on context, and makes it more effective since it discards findings that have been superseded.
Hardware wise, a TRX50 AI TOP motherboard, 7965WX CPU, RTX 4090 and x2 RTX 3060, with 128gb of 5600 RDIMM. Down the line, I might try some GPU unlockers when other folk have tested them for bugs. So about a month or two?
https://github.com/JustAResearcher/gpu-unlock-ada
https://github.com/Zagitta/gpu-unlock
https://github.com/impacifist/Windows-GPU-UNLOCK
00000000
I bought my memory and the CPU from Ebay, for prices a bit lower than at Newegg. If I could redo my GPU purchases, I would have gone with a RTX Pro 4500 Blackwell. That has 32gb of VRAM, and occupies a single double-width PCIe slot, and sips energy. The RTX 4090 is only 24gb, and takes up 1 and 1/2 of my slots.
00000000
Qwen-Flash-Next is an experimental model, the official version might come out in a month or so. If it is the same size or smaller, I will switch to that as my main driver. My current AI model feels useful to me, but clearly has raw aspects to it. Vision is a big one, since this AI can fail to see or understand details.
This is the kind of AI success story that never makes the investor decks: Your setup sounds like the future they can’t easily monetize. 💪😎✌️
I don’t use ChatGPT anymore as I’ve switched to another model but back when I did this wasn’t something they were secretly doing. It’s a feature. You can read the memories it has saved, delete the ones you don’t want it to store and IIRC it can also be turned off.
If you aren’t paying cash, you pay with your privacy. If you are paying, enable privacy features (no improving their service, no blackbox memories).
They are also subject to GDPR and similar, so make use of your legal rights, if you have any.
Or … shock, horror … don’t use them at all.
blackbox memories
I’m fairly sure you can just check what the memories are and IIRC even delete them (haven’t really used ChatGPT recently so maybe it’s changed)?
Memory summaries these days, yeah. Point & click management for ChatGPT.
Codex maintains local memory files, which is better.
Just a random sidenote to get lost in the Internet:
“Chat GPT” said out loud in French (chat, jais pété) means “cat I farted.”
You’re welcome.
Who could have seen this coming?!
“Sam Altman should be arrested” -Edward Snowden
(the whole auditorium clapped when he said that)Facebook had to stalk your clicks to build a psychological profile. OpenAI figured out people will volunteer their fears, relationships, medical worries, and political beliefs if the surveillance machine responds with: »That sounds really difficult.« The therapist is a database, and the waiting room is a data center.
Nothing new. Google, Meta, Microsoft, and (soon) Apple all probably have the same information in their databases. It is a little scary that it’s all being fed to an AI though instead of being sold to advertisers
Social media were already doing this with more relatively more primitive tech. The advent of social media mass control got me thinking. All the political and social strife. It’s only the beginning. The 20th century was marked by conventional warfare and terrifying new weapons.
The 21st century conflict is going to be something we’ve never seen before. Not unlike 20th century global conflict ushering in new forms of weaponry. We don’t know what the nuclear bomb of the 21st century will be. It will likely be something like this. Maybe we won’t be fearing an atomic bomb. Maybe people themselves will turn into violent mobs or something. What terrifying things will come out of using the world as live psychological experiments with zero ethics.
I tell two lies for every truth I say to it. My profile is probably wildly wrong.
I really just ask chatgpt why my code doesn’t work. Sadly, searches are getting worse intentionally. So, it’s either use their stupid AI, or spend hours looking for forums that used to pop up after 5 minutes of a google searching.
Kagi is with every penny
What is magic? I’ve never heard of it. I moved to duckduckgo a while back, but even that kinda sucks now.
Kagi is just a search engine without ads and works.surprisingly well; unlike ddg. Its 5 bucks a month.
You could get even better results with searxng if you have the setup and patience for it. I went kagi even as an avid self hoster because I travel a lot and tailscale would slaughter my battery if I left it on non stop
That’s the real danger with LLM-AIs IMO, not the RAM price or energy consumption.
Mass control.
Nothing to worry about, cattle!
I mean, people! Heh. Heh.
Use local AI for privacy.
