This is super cool. No longer need to fiddle with third-party LLM APIs.
For those worried that your emails contents might get leaked, I’m pretty sure they’re not using your email content to train, but only use it as context. LLMs are stateless, as soon you you start a new chat, it forgets everything from before.
That’s a different thing and they will have to ask for consent separately. I’m specifically talking about the feature of allowing Bard to access your emails/schedules/etc.
Switched to outlook but soon found out that some of your emails arent even received. That occured when I was trying to sign up to this lemmy instance. Luckily the admin was a very friendly person who understood the situation and manually approved my signup.
I’m starting to switch over to Proton. I haven’t paid for it yet but my plan is to start paying and potentially grab usernames on other sites so I can have a consistent email across any site I decide to use.
Proton unlimited was one of the best decisions I’ve ever made. Very nice vpn service that works even on arch Linux on school wifi (making sure to switch to tcp protocol), e2ee mail that I use to send confidential medical records to drs, included simplelogin premium (currently over 400+ aliases), excellent and knowledgeable customer support, and so much more. I just really love this company and their role in the FOSS, privacy, security space.
I deleted google and only have game accounts email only from simplelogin aliases. Maybe you’ll eventually realize google is a shit company and leave that environment as well.
As much as I don’t like gøøgl€, on the phone front, it’s the lesser of the two evils I can choose from currently. I probably won’t be able to outright buy my next phone before getting service and slap on a custom ROM, so I’d rather deal with the devil I know over appl€.
There definitely are options, but I’m just not sure I’ll be able to outright purchase a phone and flash a ROM like grapheneos on it. I feel as though at this point I’ll probably have to end up just upgrading through my carrier and I don’t know if they even allow you to root or flash phones from them without terminating service.
They certaintly did, and already used AI to comb through it for their monitized stalking campaign. I think the only difference now is you can now make use of this somehow too, but I only read the headline lol.
I was primarily using Gmail and Youtube Music and the switch was fairly easy. Apple Music was a better replacement than the other competitors, and Proton Mail was super easy. Paying for both makes me “not the product, but an actual customer”. Forwarded Gmail to Proton in about 30 seconds, and replaced Chrome with Firefox. Duck Duck Go isn’t a “perfect” replacement for Google Search, but it’s good enough.
It’s NOT a valid reason not to switch, but god that’s a dumb name for a product. Again, I don’t understand why I hate the name, but I do. Please help me.
Proton looks pretty good, but all the paid Google services show that paying is not enough to be seen as “not the product, but an actual customer” these days.
Curious what you decide. I feel trapped. My Gmail has been my primary account for almost 20 years. With that kind of longevity, switching would be extremely disruptive.
When I did it I just forwarded my mails to the new service (fastmail) and changed mail address whenever a mail got forwarded in. When I didn’t receive a forwarded mail for a year or so I deleted my old mail and never looked back.
If you use a password manager this get easier though, since you can just lookup where the old mail is used.
I kept Gmail, but gradually switched over to proton. Eventually, year or two later, it’s full of emails I don’t glcare about. Start now and take it slow.
I’ve heard good things about proton. Considering switching to that as well. Other than heading so many places use your old email until you update it is there any reason you did it gradually?
Try to create a new email on a privacy respecting service and start with setting up email forwarding from Gmail to the new email. Then you just kinda slowly move your accounts over for a period of months or even years. I still keep my unused Gmail forwarding just in case.
I know people who have their passwords on a google doc or email passwords. I foresee a lot of accounts getting hacked once people can crack the right prompts for the LLM.
Bard is the name of the service, they can create account specific models trained on your user data which aren’t shared with other accounts (as an extension of the base model built on public data). I’ve already read about companies doing this to avoid cross contamination. Pretty sure Google is aware of this.
But I don’t know if Google cares enough about privacy to bother training individual models to avoid cross contamination. Each model takes years worth of super computer time, so the fewer they’d need to train, the less costly.
Hmm, I thought one of the problems with LLMs was they’re pretty baked in in the training process. Maybe that was only with respect to removing information?
Yeah, it’s hard to remove data already trained into a model. But you can retrain them to add capabilities to an existing model, so if you copy one based on public data multiple times and then retrain with different sets of private data then you can save a lot of work
Their terms and privacy policy I guess. Also the possibility of data leak. I don’t think even Google would train their LLM on knowingly private data, that would be utter insanity.
Probably something similar to what they were doing back in 2010 with Google Now. Skimming data from emails and texts etc in order to give you more pertinent information with Google Assistant’s predecessor. The google now page of that time could tell me when my flight left, what gate it boarded at, if it was delayed, what airport entrance to use. It told me when my bills were due and how much. It tracked orders for me and told me when they were out for delivery or delivered. It would help me to pick a restaurant for a special occasion, direct me where to call or book for a reservation. I found it very useful and then privacy concerns basically tanked it.
We got a rebrand that did some of the same things with Google Assistant. And for a while that was really useful for a lot of the same things. But now that they’ve realised that data collection for this doesn’t net them ever increasing profits there’s a push for new better things. The new better apps and services don’t really do what people want or need. They are specifically and ever increasingly meant to funnel more data to Google and more ads to consumers. There were a lot of potentially really good useful services that this style of scraping provided. But on the other hand, they didn’t ask. They just took that info. And then saved face by sunsetting the product that people were gripping about.
If you use google services at all, google has a profile on you. Even limiting the spec of that profile by opting out of assistant and turning off a lot of their tracking doesn’t necessarily help you maintain privacy. And google services includes their app store and phones.
I wouldn’t be surprised if Alphabet has used data from all these sources all the time.
That’s actually the reason I don’t use gmail except for registering Android, or use google calender, or google search except occasionally. I have my youtube account separate from my Android, and I don’t allow any cloud services, like photo or any other storage or sync services.
The power Google can gain from using these things in combination is huge, just like Facebook influenced the 2016 US presidential election, Google/Alphabet could use this for both political and financial gains to an enormous degree, that would have been completely unheard of prior to Internet becoming widely used.
For the last three decades, tech companies have been operating in a “Ask for forgiveness, not permission” mentality.
Microsoft didn’t get such a large monopoly by playing fair. Neither did Google, Amazon, Uber, etc.
They have always said one thing, the courts found it to be inaccurate, and then they go “Oops” and pay $500 million “cost of doing business” while making 1000+ billion.
I mean, if Bard can access those things, that means Google already had access in the first place. I also don’t mind if that means I get to enjoy a better service overall.
The things I don’t want Google to know about I have separate anyways
But individual users can target specific things to get out from the llm training data. I’m more okay with Google tuning it’s spell check on my data than I am to letting anyone who knows how to ask my personality identifiable information. Even if I don’t use Google drive/docs, someone else will use Google for their contacts, and then my security questions (why do companies still make people use those?!?) are public.
I don’t mean that…I mean how pretty much every product nowadays has some sort of new “groundbreaking AI features”. AI’s got tons of practical applications, but companies are really overblowing the whole thing.
It’s already starting to, the AI groups of devs basically showed off an alpha to their CEOs, then those CEOs thought it’s game changing…while the devs were like “no it’s an alpha/pre-alpha and is really dumb”…but the CEOs rolled with it and are now finding how bad it is…so it’s been getting slowly dropped. It’s %100 a fad and has limited applications. It’s really cool tech and I have used it, but it’s not something that’s going to replace many people.
I mean who doesn't like a search engine that very convincingly and assertively lies to you? They all come off as very authoritative while also being so very wrong more often than not.
This is only true for generative and sumariziation systems that are directly consumer facing. Current AI systems have a myriad of applications for internal tooling and B2b type systems. At my work we’ve built a vast array of tools using all kinds of modern AI that have delighted our clients. Things that previously wouldn’t have been possible for the average small buisness like filling out patient information by voice.
People like to shit on AI because they think of the thing google search uses or ChatGPT but it really is changing the world just not all at once. There will be more consumer facing advencments but more than likely these will come in the forms of things like existing products that get augmented in some way to aid usability (AI tooling is fantastic at facilitating accessibility).
AI is never going away it is here to stay and it is continuing to grow and advance
While that’s technically AI… it’s really just ML dressed up. What the public thinks of AI is AGI…all you have done is describe automation at a higher level, and it’s been around for nearly two decades.
What the CEOs heard was AI…like legit AI(AGI)…what everyone who was developing it said was “here is a cool ML and general purpose automation tool”.
Lol I don’t know what meetings you have been in but there are plenty with shareholders having smoke blown up their ass… it’s why stock goes up and down…over promise under deliver.
Agreed. My lead at work wanted us to start trying/using Cursor.so (VS Code fork with AI as a builtin feature) and it’s been pretty transformative. I don’t see a lot of “hey write me a program that does x” but in my (limited) use of this, a simple “why doesn’t this function work” has been pretty amazing.
I have a feeling this is a branding issue more than anything. When you could ask google plain language questions a decade ago and get responses, that seemed amazing. This to me seems like that but more advanced and I just hope they sort out the truthiness and privacy implications. On the one hand, I want the tech to advance, on the other, I would like it to not be such a privacy nightmare.
It’s %100 a fad and has limited applications. It’s really cool tech and I have used it, but it’s not something that’s going to replace many people.
You are you trying to convince? Lol the more I hear this the more it seems like projection. You say it to comfort yourself because you’re not certain that it will but you want it to be true.
The more it gets integrated into the ecosystems that enterprise corporations use the more it will replace people. Our Data Automation engineers are currently training an in house model on company data so you can ask if for anything from the accounting spreadsheets to when the next holiday is (accounting for ACLs of course, can have accounts receivable getting access to accounts payable.)
If you genuinely think it’s a fad you most likely don’t work in an industry where it has an application or use-case, yet.
I don’t know where you’re getting this information that it’s dying out because that’s just objectively untrue. Microsoft is probably pricing out AI model Saas E5 licensing as we speak, Bing chat just got an upgrade with Enterprise Tenants recently that supposedly safeguards your company data, this shit is not going anywhere no matter what lies you’re telling yourself. It’s not replacing human oversight anytime soon but it absolutely will reduce the amount of humans behind screens at virtually every company in the next decade.
Meanwhile new jobs will open up in AI specific integration as it inevitably becomes too much for Cloud and Data engineers to handle.
“siLlY gOoGle, iTs tHe nEXt cRyPtO” just shows how little you know about it. Probably because you work in a completely unrelated industry and get your information from websites that tell you what you want to hear. Meanwhile your attempt at using AI probably amounts to breaking it’s moral safety locks and then running out of ideas. If you worked in a field that AI is being used for, and know enough to lean on its ability to work through large data sets quickly while doing simple QA work to verify its information (which takes vastly less time than searching through all the data yourself) I guarantee you that you’d find it incredibly useful.
Lol no I literally work in this field. The shit that’s being presented now is a fad. We’ve been doing ML for over a decade now, and all the cool little things you’re talking about are not new. It’s not even brand new within the last 5ish years. You’re acting like this is some brand new industry because chatgpt is the new big thing. It’s not it’s a product of the industry that’s been at this for a long time. All automation removes jobs and it’s cool, but don’t sit there and tell me it’s going to replace jobs that require thought that you cannot build into a DB. If it doesn’t have the parameters to do the job it’s not capable of doing the job.
Also I thought your Google crypto quip is great…
It’s not replacing human oversight anytime soon but it absolutely will reduce the amount of humans behind screens at virtually every company in the next decade.
Lol this is how all automation and any tech works…this shits been happening for centuries… it’s not new.
I mean, at this point I find very difficult to believe that people willingly putting their entire life in free Google accounts don’t know that they’re basically ads meat anyway. The outrage makes no sense.
This is absolutely true. Some of us were fine with the trade off in the beginning, because of how useful the services were when linked together. Google didn’t start off with everything linked like it is now. This unification of all their services with no walls or filters has been an ongoing development for more than a decade. I remember having to be on a waiting list to even get a GMail account after getting an invite from someone I knew. I’ve been around using google services since then. Back when Gmail and the calender weren’t separate apps. When YouTube wasn’t linked to Gmail or a Google account the way it is now. The landscape has changed drastically.
I wonder what will happen when proprietary Open AI LLM’s start scraping data that the companies use them don’t want scraped because it’s proprietary. IT department using LLM for code for the company wide web portal. And now schematics for a new device are scraped etc.
If you choose to use the Workspace extensions, your content from Gmail, Docs, and Drive is not seen by human reviewers, used by Bard to show you ads, or used to train the Bard model.
Add comment