Very good to see this video. Privacy is increasingly a concern, especially since they've now started arresting CEOs to get them to "voluntarily" "cooperate" with Big Brother. And to those who say, "You have nothing to be concerned about if you have nothing to hide," I reply, "Your naïvete is adorable."
Does anyone know how France got him to go to France anyway? From what i read, he didn't have any real reason to go there. It was clearly him turning himself in.
And in the words of Ed Snowdon that goes along the lines of, those that say they don't care about privacy because they have nothing to hide is like saying you don't care about freedom of speech just because you don't have anything to say.
Good video, I have a lot of concerns about AI and your work here helps me start to crystallize how I want to think about AI going forward. But I REALLY appreciate your priv/acc segment at the end, I do feel like a bit of a nutter sometimes when I try to convince my family not to share passwords over SMS text or share tons of information online. It is really nice to know that there are other weirdos like me out there. Josh Summers at All Things Secured has been similarly helpful. Thanks Naomi!
Data collecting was a concern from the beginning of web. The only problem for the collectors was, that they couldn't analyze all the available information. With AI, this changes dramatically. And only few ostracised weirdos are concerned...
I agree, but still prefer to run inference locally (as a service) on fairly janky hardware. So, this comment points to the next video… 🙈😅 I run my instance(s) as VMs on a separate network (several options) that can’t reach the internet, or the rest of my LAN. I can use it through one of several interfaces, some are “familiar” and run in a browser. RAG (feeding a model my own data) is also possible, nothing leaves the LAN. No “company” involved, as they seem to change owners (and ethics) beyond my influence. The community is great at sharing knowledge, and growing. One huge ethical issue remains, as every LLM is trained before local deployment. Even the open source models are not too transparent as to what data was used, when or to what conditions. Different topic, huge deal… Thanks as always for sharing grounded knowledge! 👍
AI years ago was called 'Expert Systems". These expert systems has a singular database and you would submit a query to have it search the data. Today with the MASSIVE amount of data, we need a better way to get through that data, so AI is here. Let AI do the searching help.
Sorry. Venice's icon looks too scary. I'm sticking with Brave. LOL! I'm going to test some AI content on there today. So glad this video popped up in my feed.
I just tried the extreme test question in Venice and it treated it about like Google would (if it were a conversational AI). So there's the same kind of schoolmarm policing of political incorrectness there you get everywhere.
Well well well, if it isn’t my Cyber Privacy AI Wifey living rent free in my mind! And I love it! I’m going to ask Chrissie Mayr if she can get you back on considering the current climate.
What I don't understand is why someone would have a conversation with AI as if they are chatting with a friend. AI is a fantastic tool for doing searches on science and tech things. Have a math question, it saves a lot of research time. What I worry about the the devices with microphones, such as your phone and those Alexia type things. Are they ease dropping? I strongly suspect the answer to that is a resounding yes. BTW: I use Brave and have been seeing the ask Leo. Didn't know what it was. I'm going to have to give it a try. That information alone made watching this worth it.
That's quite interesting with a microphone. I would be over the roof when I could use AI with a voice and be also private (so essentially your own personal assistant on the phone). Le is greag, so for example, if you use brave search engine, it will give you summary on the subject. So you can actually search for longer query rather some thought out keywords. I was able to just frame my query as a simple question and it worked. Complex query (or prompt rather) didn't work for me.
I see so much about Brave. Naomi, do you have concerns with Peter Thiel's involvement with Brave? The dude is a ground floor investor while running Palantir.
@@NaomiBrockwellTV The mouse cursor tsunami for one :) Electro Meow people would like too, I think. But perhaps just all of them? More consumer choice ;o)
The issue with a self hosted AI model is usually that the answers you get have very poor quality, especially when writing programs like in PowerShell. This problem isn't just related to programming but also to other queries.
For LLMs would be nice to get info about training them locally, using RAG, "newer" features such as Reflection.. function calling, or using them (locally) to write & work on coding projects..
This is a very timely video. We are being held back in the company from using AI for multiple data privacy and intellectual property reasons and having privacy options is great. Thanks!
Brave's Leo is nice for privacy. Also, copilot on the web does not require an account or Edge. All they get is your IP, which is fair for non sensitive use.
Since you mentioned Brave having an AI built in, I opened my Anonymous instance of Brave (everything deleted when I close that instance) and asked it what I've been doing recently on the internet. It came back with 7 things. Every one of them was wrong. Of course, it shouldn't have been able to find anything about me since I never enter anything private in that browser instance and everything's deleted at the end of the session. But, still, it pulled 7 wrong things out of it's hat and claimed they were facts.
I watch these videos of yours and look at your subscriber count and the # of views and think, "How is it that so few people watch this when someone like Mr. Beast gets zillions of views? Is he helping us in our lives to this degree?"
I can't help but worry. Not only the ethical side of training and content theft, but now that every company is stuffing AI "features" into their devices, every phone and computer...when is opting out of having it no longer an option? And how much control are these companies and governments going to have over our thoughts and what we learn if an AI has to filter it for us first? Will they be trained to avoid certain topics or info to control the narrative or current events for people? Worrisome stuff.
This is a great podcast Naomi...I have been staying away from AI because of the privacy implications but can now see that there are a number of options available...especially the Brave one.
This report fails to uncover one crucial aspect - who is paying for all of this? If the user isn’t paying, then someone else does. I am certain that they are not doing it out of the goodness of their heart. What’s the end game for them?
in the end I still consider brave a kind of scummy browser but I am intrigued by self hosted LLM style generative "AIs". I find information on that invaluable
Everyone watching this and thinking your learning something think again. She's leading you to use gov controlled apps. You will be worse off than when you started. Pro tip, she said the owners name of Venice was Eric voorhees, that's Jason's last name in the horror movie and happens to be a "tell" that gov "shenanigans" are at play. You have been warned. Hahaha!
Ideally, I would like to host my own AI model (the self-hosted option), on a machine that has no internet access (but is networked to my daily driver machine). The above would allow me to not worry about any data being sent to some data mining company, because my AI box is not able to connect to the internet. I am assuming that, for best results, I would have to periodically download the latest LLM databases, and copy them to the AI box? From where are the LLM databases obtained? And how large are they? How many 20TB drives will I need? By the way, although I trust Brave and Venice to act as a privacy proxy, and not save my queries... I do not trust bad actors with subpoena power and court order power to compel companies to keep logs and hand over data, and also issue gag orders to those companies. Self hosting solves all of the above. I never want to involve 3rd party companies for any of my business, if I can obtain the same results with my own hardware. But how fast (or slow) will my self-hosted AI system be? What kind of hardware will be fast enough to not get aggravated, performance wise? Will I need an IBM z16 mainframe? Or will something similar to a 13th generation i9 CPU with a 4,000 series GPU get the job done?
Personally I haven't used AI and so far haven't found a reason why I would need AI. There are a lot of things that bother me with AI. Companies are pushing AI on to consumers in pretty much everything but the kitchen sink. Microsoft is re-activating a nuclear power plant to power their AI and Elon Musk is becoming the single largest customer of NVIDIA. That is some crazy stuff, no? Where does that all lead? Skynet? Yeah, I watch to much TV..... 😂
Thank you for this video! I haven't touched any AI bots or chats as I personally didn't trust AI and the privacy around it. However, I've seen the benefits of others who use AI for work and getting things done, and I wanted to know if there was any way AI was able to be private. I can't wait for the next video! Still not sure if I will use AI for now, but this is definitely helpful in understanding AI as well as seeing that there are privacy options.
I’ve used Copilot and ChatGBT for resume writing and a dictionary but can this Leo AI data mine and edit/print web pages from a command? I got malware and a hacker through my Google extensions I used for those two things!
What noone tells you is that self-hosting a really powerful AI for yourself requires incredibly powerful hardware costing over 80 000 EUR... And the lesser models that run on ordinary consumer-grade computers are too weak to be competitive with ChatGPT, Gemini etc.
Wow. Thanks for all of this Great Information. Miss Brockwell is Brilliant and also reminds me of the Stunning British Ladies of the past like Jean Shrimpton and the Boyd Sisters. Win/Win
I wonder if that's enough, though. Even if you're self hosting, that doesn't solve the problem that new computers with new operating systems are integrated with their own AI hardware and software to "help" you. Is Microsoft or Apple looking and remembering everything you type or do - or for space reasons consolidating it but still knowing everything? Even if it's not going back to their servers, all that info might be on your device with now way of cleaning it out short of a sledge hammer. You don't have to worry about testifying against yourself. Your laptop or mobile phone will do it for you since it's always there.
Ms. Brockwell: A subpoena is a subpoena. If someone with "home" AI is subpoened, that is not a tad different than a "hub" as you call it. Please stay away from the legalities, or talk with an attorney to get a better understanding. I'm not an lawyer. I was a cookbook author.
Personally I think the whole term "AI" is so overused nowadays. For example the whole Samsung AI battery learning pattern is basically something that existed for a quite a while without AI. Same with Adobe and many other shitty services. But guess what, back in my days, it was called a Generator. For example there was a story generator back in 2009 which I used a few times for school and it was basically the same like ChatGPT. Same with picture generators which were on Facebook when all the boomers and kids started using apps like "Child name generator" or "how successful will you be in x years" or that Farmville game 😂 Humans are not ready for the real AI and the attempts of the Tesla Robot and Tesla self driving taxis just navigates the humanity into one direction and it'll most likely end up with something like Skynet with killer robots or the Tradis from the old Dr. Who (or however the robots were called, which were screaming "Exterminate" before shooting a little laser beam 😂 )
Very good to see this video. Privacy is increasingly a concern, especially since they've now started arresting CEOs to get them to "voluntarily" "cooperate" with Big Brother. And to those who say, "You have nothing to be concerned about if you have nothing to hide," I reply, "Your naïvete is adorable."
Does anyone know how France got him to go to France anyway? From what i read, he didn't have any real reason to go there. It was clearly him turning himself in.
And in the words of Ed Snowdon that goes along the lines of, those that say they don't care about privacy because they have nothing to hide is like saying you don't care about freedom of speech just because you don't have anything to say.
What CEOs have been arrested for the reasons you claimed? I'd really like to read about it.
It's not that my conduct and intentions questionable. It's that other people's are.
@Damons-Old-Soul was invited to have diner with la pen or something. Someone up there.
The Leo summaries for Brave search are great. It also shows which articles it got each part of the answer from.
Good video, I have a lot of concerns about AI and your work here helps me start to crystallize how I want to think about AI going forward. But I REALLY appreciate your priv/acc segment at the end, I do feel like a bit of a nutter sometimes when I try to convince my family not to share passwords over SMS text or share tons of information online. It is really nice to know that there are other weirdos like me out there. Josh Summers at All Things Secured has been similarly helpful. Thanks Naomi!
Yay here's for solidarity with all us weirdos :)
Data collecting was a concern from the beginning of web. The only problem for the collectors was, that they couldn't analyze all the available information. With AI, this changes dramatically. And only few ostracised weirdos are concerned...
It will be interesting to see how many people react to this by saying, Why should I care? I've got nothing to hide.
They can watch me in the bathroom 🚽 since I got nothing to hide and did nothing wrong
Prfect timing as I've been concerned more and more about this issue - Amazing! Thank you.
good to know and yes people seem to not take their privacy seriously enough on so many things,
I agree, but find it hard to use them, I often get annoyed when I use them. I hope they will get truly competitive soon. We need this!
I agree, but still prefer to run inference locally (as a service) on fairly janky hardware. So, this comment points to the next video… 🙈😅
I run my instance(s) as VMs on a separate network (several options) that can’t reach the internet, or the rest of my LAN. I can use it through one of several interfaces, some are “familiar” and run in a browser. RAG (feeding a model my own data) is also possible, nothing leaves the LAN.
No “company” involved, as they seem to change owners (and ethics) beyond my influence. The community is great at sharing knowledge, and growing.
One huge ethical issue remains, as every LLM is trained before local deployment. Even the open source models are not too transparent as to what data was used, when or to what conditions.
Different topic, huge deal… Thanks as always for sharing grounded knowledge! 👍
Would love to see a vid on how to have a private real physical address (where you do not live), not PO Box or equivalent. Ty! ❤
I'm having serious issues atm as USAA refuses to return my VA check, almost a month now. They R sooo broken!
You can use your post office's street address. Get a proper PO Box not a private one. I do this. Have for half a decade. :)
USAA is awful. Threaten to sue, they'll cave. FAST. Get a credit union.
@@2rx_bni What is a "proper PO Box" ? Thank you!
@@2rx_bni Some will some will not. But ty
Wow! This is a very interesting turn of events. Thank you so much for this post.
So educational. Thank you, Naomi!
AI years ago was called 'Expert Systems". These expert systems has a singular database and you would submit a query to have it search the data.
Today with the MASSIVE amount of data, we need a better way to get through that data, so AI is here.
Let AI do the searching help.
Research Requires Coffee ☕️
Thanks so much for your support!
Very interesting, thank you. I have now installed Brave. Thank you.
What we really need is a video about AI on phones as they become more and more common, especially with Apple fast approaching iOS 18.1
Sorry. Venice's icon looks too scary. I'm sticking with Brave. LOL!
I'm going to test some AI content on there today. So glad this video popped up in my feed.
I just tried the extreme test question in Venice and it treated it about like Google would (if it were a conversational AI). So there's the same kind of schoolmarm policing of political incorrectness there you get everywhere.
Venice allows you to use larger private models that is one benefit of Venice. Just a heads up but the icon is wierd looking.
@@cryptowealthonyt Larger private models -- I'm lost.
I just wanted to say, thank you!. You are giving a great public service, even though, most people don't realize that.✌️
Thank you, thank you, thank you 👍 Just the info I was looking for... So important topic.
yes, this is a very important conversation
Well well well, if it isn’t my Cyber Privacy AI Wifey living rent free in my mind! And I love it! I’m going to ask Chrissie Mayr if she can get you back on considering the current climate.
Makes AI Brockwell 1.0…and she’s sassy like Cortana! 😜
What I don't understand is why someone would have a conversation with AI as if they are chatting with a friend. AI is a fantastic tool for doing searches on science and tech things. Have a math question, it saves a lot of research time.
What I worry about the the devices with microphones, such as your phone and those Alexia type things. Are they ease dropping? I strongly suspect the answer to that is a resounding yes.
BTW: I use Brave and have been seeing the ask Leo. Didn't know what it was. I'm going to have to give it a try. That information alone made watching this worth it.
That's quite interesting with a microphone. I would be over the roof when I could use AI with a voice and be also private (so essentially your own personal assistant on the phone).
Le is greag, so for example, if you use brave search engine, it will give you summary on the subject. So you can actually search for longer query rather some thought out keywords. I was able to just frame my query as a simple question and it worked. Complex query (or prompt rather) didn't work for me.
Thank you Naomi from Australia!
I see so much about Brave. Naomi, do you have concerns with Peter Thiel's involvement with Brave? The dude is a ground floor investor while running Palantir.
NBTV essentials 💯
My brain immediately swaps out LLM for MLM
Thanks Naomi, always great videos
Nice that there are privacy focused AIs now!
Regarding the shop: would be cool if you'd have zip-up hoodies of more than one design on there as well!
ooh what design do you think I should add?
@@NaomiBrockwellTV The mouse cursor tsunami for one :) Electro Meow people would like too, I think. But perhaps just all of them? More consumer choice ;o)
Great content. Brave installed. Subscribed. Good work.
The issue with a self hosted AI model is usually that the answers you get have very poor quality, especially when writing programs like in PowerShell. This problem isn't just related to programming but also to other queries.
Thanks for doing this kind of videos!
Thanks for these videos, are so useful ❤
Because privacy, matters ❤
Both are great resources.
For LLMs would be nice to get info about training them locally, using RAG, "newer" features such as Reflection.. function calling, or using them (locally) to write & work on coding projects..
I think that if all LLMs were required to be open source somehow, that could go a long way to improve privacy while working with AI tools.
Thank you for this very important topic. I use Venice Ai now and I feel better with my searchs I do.
Great video. Thanks!
I am having a gpu installed into my system. I like running models in ollama or lm-studio. Very great applications both.
This is a very timely video. We are being held back in the company from using AI for multiple data privacy and intellectual property reasons and having privacy options is great. Thanks!
Thanks Naomi.
Love your work!
Keep it coming!
Brave's Leo is nice for privacy. Also, copilot on the web does not require an account or Edge. All they get is your IP, which is fair for non sensitive use.
Tried Leo, asked some ''2024 sensitive'' questions, never seen it again.
@@violet_world9385 Yeah, Mistral and Claude are bad for real world research. I only use it for writing and coding.
Since you mentioned Brave having an AI built in, I opened my Anonymous instance of Brave (everything deleted when I close that instance) and asked it what I've been doing recently on the internet. It came back with 7 things. Every one of them was wrong. Of course, it shouldn't have been able to find anything about me since I never enter anything private in that browser instance and everything's deleted at the end of the session. But, still, it pulled 7 wrong things out of it's hat and claimed they were facts.
Even when it's local, there is gonna be some people who will say that AI is terrible.
Wow so basically there integrated Open web ui and Ollama into brave. Nice
Such quality Naomi
What about Ollama? I've been using it for offline AI on my laptop (modest 8GB Ram 11th gen intel).
yep! Next video in the series is about that!
I use brave religiously. Every once a while I use Leo. Can Leo be used offline?
20:59 is so true 👍
I watch these videos of yours and look at your subscriber count and the # of views and think, "How is it that so few people watch this when someone like Mr. Beast gets zillions of views? Is he helping us in our lives to this degree?"
Excellent job! Keep up the good work 👍
I can't help but worry. Not only the ethical side of training and content theft, but now that every company is stuffing AI "features" into their devices, every phone and computer...when is opting out of having it no longer an option? And how much control are these companies and governments going to have over our thoughts and what we learn if an AI has to filter it for us first? Will they be trained to avoid certain topics or info to control the narrative or current events for people? Worrisome stuff.
That's precisely how they're being used currently and also in the future as well.
This is a great podcast Naomi...I have been staying away from AI because of the privacy implications but can now see that there are a number of options available...especially the Brave one.
This report fails to uncover one crucial aspect - who is paying for all of this? If the user isn’t paying, then someone else does. I am certain that they are not doing it out of the goodness of their heart. What’s the end game for them?
Brave and venice both have a freemium tiered model
in the end I still consider brave a kind of scummy browser but I am intrigued by self hosted LLM style generative "AIs". I find information on that invaluable
Why is brave a scummy browser lol brave is the best
Did it ever tell you that it's looking for a guy named John Connor or his mother Sarah?
Yeah, I showed him how to use google maps. He told me he WAS google maps. Weird. Seemed like a nice guy.
LM Studio is great
Good video!
Great and awesome video. Can't hardly wait for the next video.
Id like to see if self hosted ai has a lower power comsumption. Im just as concerned with the environmental impacts as i am the privacy concerns
Oh no are you feeling ok Naomi? Sounds like you have a cold. I hope you feel better!
Oh yeah I was crazy ill when I filmed that :( Sorry if it sounded bad!
Australian elegance on full display as always. Thank you for another upload, Naomi!
She doesn’t live in Australia
What's the best email to use? And message plus is being discontinued any suggersting for text msg
15:21 I wish 15 pages was considered long for a privacy policy, lol.
How is brave paying for the compute needed for those llms?
❤ from india 🇮🇳
👋 Hello my friend!
So with Venice AI you can navigate Venice’s local canals while they give you the benefit of a far canal?
Everyone watching this and thinking your learning something think again. She's leading you to use gov controlled apps. You will be worse off than when you started. Pro tip, she said the owners name of Venice was Eric voorhees, that's Jason's last name in the horror movie and happens to be a "tell" that gov "shenanigans" are at play. You have been warned. Hahaha!
Takk!
Thanks so much for your support!
Ideally, I would like to host my own AI model (the self-hosted option), on a machine that has no internet access (but is networked to my daily driver machine).
The above would allow me to not worry about any data being sent to some data mining company, because my AI box is not able to connect to the internet.
I am assuming that, for best results, I would have to periodically download the latest LLM databases, and copy them to the AI box?
From where are the LLM databases obtained? And how large are they? How many 20TB drives will I need?
By the way, although I trust Brave and Venice to act as a privacy proxy, and not save my queries... I do not trust bad actors with subpoena power and court order power to compel companies to keep logs and hand over data, and also issue gag orders to those companies.
Self hosting solves all of the above.
I never want to involve 3rd party companies for any of my business, if I can obtain the same results with my own hardware.
But how fast (or slow) will my self-hosted AI system be? What kind of hardware will be fast enough to not get aggravated, performance wise?
Will I need an IBM z16 mainframe?
Or will something similar to a 13th generation i9 CPU with a 4,000 series GPU get the job done?
Personally I haven't used AI and so far haven't found a reason why I would need AI. There are a lot of things that bother me with AI. Companies are pushing AI on to consumers in pretty much everything but the kitchen sink. Microsoft is re-activating a nuclear power plant to power their AI and Elon Musk is becoming the single largest customer of NVIDIA. That is some crazy stuff, no? Where does that all lead? Skynet? Yeah, I watch to much TV..... 😂
Great hair Naomi :-).
Thank you for this video! I haven't touched any AI bots or chats as I personally didn't trust AI and the privacy around it. However, I've seen the benefits of others who use AI for work and getting things done, and I wanted to know if there was any way AI was able to be private. I can't wait for the next video! Still not sure if I will use AI for now, but this is definitely helpful in understanding AI as well as seeing that there are privacy options.
So how do they train their data?
build your own ai tool instead of using exiting tool, always better
good info.. but I still don't like the crypto crap that is integrated with Brave..
you've not mentioned Kagi,Naomi; it is privacy concerned and the company is AI based!!
Beep boop I'm a 🤖
ME TOO!! A bot that is, not a robot but you know a bot! Shall we claim all their base?
Hi there fellow bot! Were you here for all their base? Doesn't matter because it now belongs to us! Haha beep boop ha bop!
I’ve used Copilot and ChatGBT for resume writing and a dictionary but can this Leo AI data mine and edit/print web pages from a command? I got malware and a hacker through my Google extensions I used for those two things!
what have to do, carry our gadgets in rfid secure bags, aluminium bags ?? ..
priv/acc!
yesssssss!!!!
Is that "Anatomy of the State" in the background? Automatic subscribe
:)
I'm not hiding from anyone. I'm hiding from everyone.
What about jailbreak dose it work?
What noone tells you is that self-hosting a really powerful AI for yourself requires incredibly powerful hardware costing over 80 000 EUR... And the lesser models that run on ordinary consumer-grade computers are too weak to be competitive with ChatGPT, Gemini etc.
Not necessarily. Check out video on the topic from Dave’s garage.
Wow. Thanks for all of this Great Information.
Miss Brockwell is Brilliant and also reminds me of the Stunning British Ladies of the past like Jean Shrimpton and the Boyd Sisters. Win/Win
Thanks for the video. For people: LM Studio for the LLMs and Stability Matrix for Images for the win
I just started LM Studio based on your comment and it's great. Installing models is easy. Haven't tried SM yet but I'll have a look later
So how is this all going to work with Apple 16?
Our phone does that now. Isnt that bixby ?
Love your efforts, Naomi!
thank you so much for your support! It's really appreciated!
What about keylogging m
Where is the video to install Leo locally?
so informative, helpful, exhaustive beautiful lady
New rule: everyone using AI should spend a few minutes each day telling it lies.
I wonder if that's enough, though. Even if you're self hosting, that doesn't solve the problem that new computers with new operating systems are integrated with their own AI hardware and software to "help" you. Is Microsoft or Apple looking and remembering everything you type or do - or for space reasons consolidating it but still knowing everything? Even if it's not going back to their servers, all that info might be on your device with now way of cleaning it out short of a sledge hammer. You don't have to worry about testifying against yourself. Your laptop or mobile phone will do it for you since it's always there.
It's possible I've just experienced love at first sight
Ms. Brockwell: A subpoena is a subpoena. If someone with "home" AI is subpoened, that is not a tad different than a "hub" as you call it. Please stay away from the legalities, or talk with an attorney to get a better understanding. I'm not an lawyer. I was a cookbook author.
If someone wants your data on your home device, they can get a specific warrant, not a general subpoena.
Do you have a paypal account for donation?
We have many options including Paypal! Thanks so much for asking! NBTV.media/support
Personally I think the whole term "AI" is so overused nowadays. For example the whole Samsung AI battery learning pattern is basically something that existed for a quite a while without AI.
Same with Adobe and many other shitty services.
But guess what, back in my days, it was called a Generator. For example there was a story generator back in 2009 which I used a few times for school and it was basically the same like ChatGPT. Same with picture generators which were on Facebook when all the boomers and kids started using apps like "Child name generator" or "how successful will you be in x years" or that Farmville game 😂
Humans are not ready for the real AI and the attempts of the Tesla Robot and Tesla self driving taxis just navigates the humanity into one direction and it'll most likely end up with something like Skynet with killer robots or the Tradis from the old Dr. Who (or however the robots were called, which were screaming "Exterminate" before shooting a little laser beam 😂 )
How about DuckDuckGo they have an ai Is it private like brave
Duck Duck Go isn't as private as they say they are. Brave is really about protecting our privacy.
Duck duck = micro soft
I try to stay away from AI. Rather give out what information by my own choice.