Share your feedback on the AI services experiment in Nightly

[asafko]

Hi folks, 

In the next few days, we will start the Nightly experiment which provides easy access to AI services from the sidebar. This functionality is entirely optional, and it’s there to see if it’s a helpful addition to Firefox. It is not built into any core functionality and needs to be turned on by you to see it. 

If you want to try the experiment, activate it via Nightly Settings > Firefox Labs (please see full instructions here). 

We’d love to hear your feedback once you try out the feature, and we’re open to all your ideas and thoughts, whether it’s small tweaks to the current experience or big, creative suggestions that could boost your productivity and make accessing your favorite tools and services in Firefox even easier.

Thanks so much for helping us improve Firefox!

[Mardak]

These Nightly builds have the new sidebar experience with vertical tabs in development along with the chatbot experiment. We're still working on integrating these features, but you can have both active today.

 

[aminought]

Sidebery > native vertical tabs. Also:
1. Each sidebar extension should have their own width. It is very inconvenient to look at the result of the AI Chatbot in a small panel in which I usually have Sidebery.
2. Some sidebar extensions, like AI Chatbot, should be able to open in the overlay so that I can return to Sidebery without additional actions.
3. A setting is needed to select the language in which the chatbot should respond.

Thank you!

[Iron56716]

I honestly like the AI Chatbot integration. I'm hoping Google AI Studio will be added officially.

It'd be even better if the sidebar could get official web page integrations like Edge or Floorp, That feature is very handy.

[Mardak]

Curious, is there a particular functionality you like about Google AI Studio? The additional settings for Gemini model choice? Model tuning? Here's using it as a custom provider, but I had to manually copy/paste the prompt:

 

[Iron56716]

There's quite a lot of things I like about Google AI Studio. Firstly, the ability to completely turn off all the safety filters, it's much easier to chat with the AI without them. Secondly, like you mentioned - the Gemini Model choice. I can change the model to any different one right in the conversation without the need to open a new chat. And of course, there's some extra perks like Custom Instructions, Model tuning, etc. Last but not least, AI Studio allows the use of Gemini 1.5 Pro without the subscription (unlike the regular Gemini websites. Although, the AI Studio version of 1.5 Pro does have a 50 turn rate limit I believe).

[everentropy]

Why are you even using Firefox if you're just giving your information to Google products anyway...

[mcortt]

AI in sidebar is great. However, please give us some sort of keyboard shortcut to open it.

[Mardak]

Does it need to be a keyboard shortcut or would a dedicated button to click be good for now? There's a lot of existing keyboard shortcuts, so having one for each sidebar might be too many, but could a keyboard shortcut to toggle / reopen the last opened sidebar work too?

[mcortt]

The main goal of a web browser is to browse the web, not browse the browser UI, which is what I feel a specific button accomplishes. The button would work (the sidebar toggle button is what I currently use), but there is no keyboard shortcut to toggle it currently that I am aware of. Personally, I would prefer a keyboard shortcut for the AI chat specifically, but I am also of the thought that most things should have keyboard shortcuts, or at least be able to be assigned a keyboard shortcut (a la extension shortcuts). Though, I understand where that could be a lot of work.

My preferences in order:

• Assignable keyboard shortcut to AI, synced tabs, etc
• Keyboard shortcut for sidebar toggle
• Button to open it

For now I would settle and be happy with a keyboard shortcut to toggle the sidebar to its last state.

[ppm1337]

I disagree. As a user, I want to have a consistent UX. In my mental model of using Firefox I want to be able to access all features that I consider useful via a keyboard shortcut. The moment I need to regress from my default behaviour of keyboard navigation, a feature becomes less appealing.

Also, most other sidebars are accessible via keyboard (Ctrl + H for history, Ctrl + B for bookmarks, ...) why suddenly have a sidebar that's deviating from this scheme?

[jtha]

I agree with the others, and frankly I would not use this if it was not a shortcut key away. Even less inclined to use this if I have high resolution / widescreens, this just adds friction. Ideally in general users could set the shortcut keys themselves if they want to, and assign to which keys. I bet you most of the shortcut keys aren't used anyway, so instead of having them all pre-assigned and firefox deciding which gets shortcuts which don't, leave it to the users to decide, if they want to.

[benlk]

Making keyboard shortcuts configurable to the user would solve a lot of concerns here: people can assign what works for them, and if they don't want to unintentionally invoke the AI tool, they can unassign the shortcut. See https://connect.mozilla.org/t5/ideas/customizable-hotkeys/idi-p/4979

[Henrymamar]

The AI in the sidebar is a fantastic feature for quick access to information, but adding a keyboard shortcut would make it even more convenient. A shortcut would allow users to open the sidebar instantly, enhancing productivity and streamlining workflow. Hopefully, this functionality can be added soon for an even smoother user experience!

[BiORNADE]

You can put ChatGPT or any AI Chatting service into a sidebar, which makes the features useless and sort of a filler but... Actually, the only time that AI Chatbots in a browser was a big deal was when Microsoft Edge had BingAI, which was a free version of GPT 4.0 FE (with some improved enhancements), but it didn't last much until MS got Copilot (BingAI) in almost everywhere, so you can use it on Windows 7 now (sarcastically saying)

To test the waters with sidebar, using a known website (from time to time) is a nice idea, so the Firefox Team are able to detect the issue easier and deliver a fix earlier... But a native AI feature that works just like a sidebars is not that of an idea

But here's a nice thing; you can actually access everything just like watching a gecko video and commenting on this very forum using a simple technique.
Some more contrast in the lines, some more tests, some polish from the community, the final testing, then the final product. The technique could also extend to PWAs (the most Kudoed question still on review)

Looking ahead to an answer

Kind Regards,
BiORNADE

[Mardak]

I'm guessing you used Google Gemini to navigate the sidebar to another page such as this Connect discussion? You could more directly do that by setting a custom provider url, but that's not the main use case of this feature.

[BiORNADE]

It almost works like a Sidebar, so you can just access anything with little to no issues.

Wouldn't make sense to have ChatBot option when Sidebar gets complete, but maybe some little additions that you can make like:

[to provide ChatBOT Features] -> IF<TitleGPT is the only website marked as CHATBOT (1)> or IF<TitleGPT is open out of all marked ChatBots (2+)> == (Chatbot actions shown for TitleGPT only)
<Otherwise, if there are no Chatbots added on the sidebar> == [all features of Chatbots must be disabled]

Atleast for me, adding Chatbot (and an option to mark as) to a sidebar feature instead of a separate thing, would be better. Just don't make it so bloated, but just do some, like, simple stuff

[Frisk]

Hello!

With introduction of even more AI services to Firefox I wanted to express that to me it does seem like Firefox is missing with development of features. This sentiment is echoed by a lot of people in my social bubble of technologists, ethicists and other people with same priorities as what one could think are values which Firefox was built on.

Those concerns are in my opinion very valid. The machine learning models have shown to be unreliable - just some of the recent examples from AI products made by large corporations: pointing users to eat pizza with glue, providing false information about just about anything. There are a number of ethical issues yet to be resolved with usage of AI, from its intense usage of computing resources that adds up to electric grid demands. Through privacy and possible copyright violations of datasets that power the models. To an entire bag of other issues monitored by excellent resources such as AIAAIC repository. 

AI/Machine learning is an amazing field with many likely applications, and yet, its recent rise to fame is characterized by failures and issues in many implementations. Personally I often don't even see whether the application of AI is truthfully necessary, in many cases a human would do a more trustworthy and fast task of information gathering than a large machine learning model.

When we compare current state of "AI", does it reflect what Firefox stands for? Does it reflect Mozilla's principals?

Let's compare.

• Principle 2

  The internet is a global public resource that must remain open and accessible.

LLMs are known for being a black box. Depending on our definition of "open and accessible", LLMs can be a very free resource or a completely inaccessible black box of math.

• Principle 4

  Individuals’ security and privacy on the internet are fundamental and must not be treated as optional.

It's clear that LLMs pose a privacy risk to Internet users. LLMs pose risk in at least two ways - because the data they are trained on sometimes contains private information due to negligent training process. In this case users of a learned model can possibly access private information. The second risk is of course usage of 3rd party services that may use information to infringe on privacy of users. While Mozilla in blog assures that "we are committed to following the principles of user choice, agency, and privacy as we bring AI-powered enhancements to Firefox", it's unclear how supporting such services as "ChatGPT, Google Gemini, HuggingChat, and Le Chat Mistral" helps protect Firefox user privacy. Giving users choice should not compromise their safety and privacy.

• Principle 10

  Magnifying the public benefit aspects of the internet is an important goal, worthy of time, attention and commitment.

In my opinion in the process of designing AI functionalities on top of Firefox there was no evaluation on how those functionalities can benefit the public. There are a number of issues as mentioned above with the LLMs, they can be dangerous and work in detriment to users. Investing and supporting in technology of this type may lead to terrible consequences with little actual benefit.

In addition Mozilla claims:

• We are committed to an internet that elevates critical thinking, reasoned argument, shared knowledge, and verifiable facts.

I argue that AI models are the opposite to that. AI output is not verifiable. They are working against sharing knowledge by making seemingly accurate information that turn out to be false.

I ask Mozilla to reevaluate impact of AI considering all of those points. I ask on behalf of myself as well as many users that I see on Fediverse being greatly worried and frustrated with AI changes added on top of Firefox. There is certainly a lot of potential greatness that could be done with AI, but those steps must be taken responsibly.

[Mardak]

Thanks for the thoughtful feedback that many might have trouble writing up. Specifically about open and accessible, we've built the Firefox feature with transparency and interoperability with any webpage. This encourages diversity in the AI ecosystem, so perhaps competition and innovation will help make these systems less of a black box and improve on many other aspects like privacy as well.

Not everyone has your skill nor background to participate in these discussions. There's potential for these free and readily available AI chatbots to help many others universally and inclusively share their ideas and concerns. Indeed, there's a lot of potential greatness in these tools as well as responsibility in how they get used.

[phreaknerd]

And this answer is a perfect example for meaningless marketing-speech and not taking serious responsibility. By honoring the commenters competency and repeating his/her arguments you are exactly saying nothing.
What will you really do to take care of those issues?
When will you do it?
How will you keep us updated about the consequences?
Us - the users and advocates of your product - we are the multiplicators: explaining, promoting and defending it - if you allow us to. The latest developments with AI and ad-tracking is chilling my motivation extremely to continue with this...

[Mardak]

Are you asking about something in Firefox where you're looking for something more open and private than running a LLM on your own device? If more broadly, are you looking for something like https://blog.mozilla.org/en/mozilla/internet-policy/mozilla-urges-federal-privacy-law-for-ai-develop...

[JHeuser]

No. They want you to honor your company's foundation and remove this, like everyone else here. Read the room.

[MageBunkshelf]

Your responses throughout seem to indicate that Mozilla's stance is they're going to just forge on ahead anyway despite 99% of the userbase here saying don't.

[zedisnotdead]

---

@Mardak wrote:

Thanks for the thoughtful feedback that many might have trouble writing up. Specifically about open and accessible, we've built the Firefox feature with transparency and interoperability with any webpage. This encourages diversity in the AI ecosystem, so perhaps competition and innovation will help make these systems less of a black box and improve on many other aspects like privacy as well.

Not everyone has your skill nor background to participate in these discussions. There's potential for these free and readily available AI chatbots to help many others universally and inclusively share their ideas and concerns. Indeed, there's a lot of potential greatness in these tools as well as responsibility in how they get used.

---

I would like to apologize in advance for what I am about to say, but this is a load of marketing nothing-speak. What you are currently doing is just a solution looking for a problem. To "help many others universally and inclusively share their ideas and concerns", we need nothing more than to let them, and just learn to understand them better.

Furthermore, LLMs will never be "less of a black box" by design. What we do is tweak the way neural networks learn, not what they do - we can only pretend we can do that with prompt """engineering""". Simply offloading the computation to my own computer doesn't break the black box open, it just moves it to my house.

LLMs are a hype train-powered future liability, a ticking legal time bomb, and as it was mentioned by many other commenters - has no place in Firefox, nor in any other Mozilla products.

If this *has to* be implemented, extract the capabilities to create a new button in the sidebar and embedding websites out (because that's a good idea) and make it an extension.

[qu1j0t3]

It's worse than "a solution looking for a problem". 

It's a non-solution that is creating and exacerbating very serious problems. It's unacceptable. Mozilla will destroy their brand if they continue to humour LLM and generative garbage.

[Zefram]

You can really see the frantic race to try and make this technology ubiquitous and indispensable before the law can catch up. The think if they can make this into something "too big to fail" then the powers that be will just throw up their hands.

[mega]

Did an AI write this?

[JPLeBreton]

---

@Mardak wrote:

, so perhaps competition and innovation will help make these systems less of a black box and improve on many other aspects like privacy as well.

---

Without putting aside all the social-political objections I have to LLMs (which others have articulated many of well in this thread), this is pure "faith in markets" wishful thinking of the kind that has led Mozilla down dead end alleyways in the past. By building conduits from the for-profit LLMs into Firefox you are strengthening the position of those players, which already have large advantages over open source LLMs. Even if they "have no moat", simply insisting that open source will win in this space when the most fundamental terms of the game have been set by those with the most capital to wield is naive and ahistorical.

Strategically this feature seems like it is more about Mozilla executive FOMO, avoiding getting left behind, than doing the most good for the most users and reaching new users who might be fleeing the Chrome monoculture. Right now Firefox needs to restore trust, and LLMs are a technology with major trust concerns on multiple fronts. As you can see in this thread, giving OpenAI (a company with notoriously untrustworthy leadership) real estate in your core browser functionality is creating rather than answering privacy questions, and saying "well if the privacy concerns are real then the market will sort it out"... oof. The market has provided ample evidence that the best-capitalized players will utterly crush Firefox in any way they can. So this seems like preemptively conceding defeat to what is increasingly clearly a bubble, and the only viewpoint I can imagine to whom it's not is one that just thinks LLMs are neat and cool and the (unavoidable) future, with no actual downsides... which is an embarrassing position for champions of the open web. The world is something we make.

[xenianth]

ignoring the extensive and knowledgeable criticism to take jabs at others who didn't write in the same way by shilling said chatbots. this is patronizing and disrespectful to both the OP and others who don't write in a way you like.

[CharlieBo]

Rude.

[doodlemancy]

The dismissiveness and arrogance with which you responded to this very smart, very thoughtful person, who gave you their time and a lot of things to think about, is honestly sickening.

[Sintanity]

"Thank you for your actually thoughtful feedback! Allow me to read none of it, take none of it in, and reply by implying you agree that AI can be good." Bruh

[jseakle]

Hey Mardak,

Reading your responses here, it's clear that you've put a lot of love into this feature and that your heart is in the right place. I know you're getting a lot of angry pushback and it's probably causing you to go on the defensive. People are reacting strongly because they are *scared*, and maybe in some cases that makes them think you are evil, but you're not. You're just coming from really different places.

So I want to ask you really genuinely & kindly to take a step back and consider the bigger picture with fresh eyes. Chatbots are cool and will definitely help people. But at the same time, with a userbase the size of Firefox's, it's not a question of whether it will say something deeply racist, dangerous, or misleading, nor even of when — it's a question of how many times per day and does Mozilla get sued over it and whether anyone dies as a result. No matter how well you try to filter your output, some child is going to be instructed in how to use the stove (even correctly!) and end up causing a gas fire. A Palestinian is going to be told that their people are not experiencing a genocide. Someone making a website is going to given code with a SQL injection in it and lose everything, including maybe other people's credit card information or medical records. Even the attempts to prevent issues like these create more issues — by trying to enforce "neutrality" in responses, you end up explicitly endorsing the status quo on issues where Mozilla has no business endorsing anything.

This tool just isn't ready yet. It can do so many cool things, when used by people who know what they are signing up for and take responsibility in seeking it out on their own. But putting it inside a web browser where everyone will see it without asking, including children, the elderly, the new-to-computers, etc. is simply irresponsible. Please hold off for now.

Thank you.

[Obligada]

Respectfully, I’d argue that people who object to this ai feature are not scared. We are angry. We are frustrated at seeing this pointless, actively harmful technology being added to a web browser that we believed we could trust with our privacy. In a general sense yes, I am scared of AI’s impact on the environment and how practically everyone involved in pushing it seems to ignore or laugh off these concerns. This is actively making the world a worse place and for what? 

I’ve read as many replies as I could, and I’ve yet to see an explanation as to why this is being implemented, let anyone anything resembling an actual answer to our objections and concerns. Calling it “scared” seems to imply our rejection of this feature is irrational. It isn’t. So many people have eloquently argued against it and we get nothing but silence. 

[dhasenan]

---

@Mardak wrote:

Not everyone has your skill nor background to participate in these discussions. There's potential for these free and readily available AI chatbots to help many others universally and inclusively share their ideas and concerns.

In other words, you want someone who has a bad feeling about something to consult an LLM to generate some text objecting to the idea. The text will not reflect their knowledge (they don't have it) and it won't reflect the LLM's knowledge (it doesn't actually know things, it just repeats text similar to what it's seen). The LLM draws on text that was seen in similar contexts, but there might be salient differences (for instance, a different browser might send text to an online service for translation whenever you highlight it on the page; if there was a lot of discussion about that, the LLM is likely to draw from those discussions in its generated text).

This is not productive. I'm quite certain you do not in fact want to read hundreds of comments like this. It is much more productive for someone to realize they don't know that much about the situation and either not comment at all or educate themselves before commenting.

I hope you are being paid handsomely to promote this, because otherwise it would just be sad.

[lucideer]

> Thanks for the thoughtful feedback that many might have trouble writing up.

Frisk clearly put a lot of time & effort into that detailed & considered post & you've completely disrespected their thoughtful feedback by replying with disingenuous & meaningless marketing buzzwords. Your second paragraph is also extremely condescending - have you considered the possibility that may not have the "skill nor background to participate in these discussions".

[everentropy]

So what I'm hearing from this response is you only want to listen to users who can give you a small essay, and not listen to he overwhelming sentiment that people don't want this. Really makes me feel valued as a user.

[Vanastar]

Why does the AI ecosystem need greater diversity? What does that look like, and why do you think it is a good thing?

[MageBunkshelf]

---

@Mardak wrote:

Not everyone has your skill nor background to participate in these discussions. There's potential for these free and readily available AI chatbots to help many others universally and inclusively share their ideas and concerns. Indeed, there's a lot of potential greatness in these tools as well as responsibility in how they get used.

---

So this tool is supposed to be for people with not much knowledge on technology? The kind of people who won't know that, as you yourself said elsewhere here, the AI responses are "correctly" not to be trusted? You yourself said that here, and now you're saying the intended demographic is people who lack the skill and background understand it?

This is why it should be an extension and not built into the browser itself: you should not be throwing a mass audience who doesn't have the skill or background to understand what AI is or what it does at a machine that hallucinates false "facts", plagiarises scraped data, and is not to be trusted. The reason you're getting this backlash is because this reads incredibly strongly as corporate selling out the browser at the expense of the actual Firefox userbase because the sequel to NFTs wants some advertising. Frankly this debacle is a good-will nightmare for Mozilla, and I'm shocked you still went ahead with it after the response you got back in June.

[Mallow]

---

@Frisk wrote:

Hello!

With introduction of even more AI services to Firefox I wanted to express that to me it does seem like Firefox is missing with development of features. This sentiment is echoed by a lot of people in my social bubble of technologists, ethicists and other people with same priorities as what one could think are values which Firefox was built on.

Those concerns are in my opinion very valid. The machine learning models have shown to be unreliable - just some of the recent examples from AI products made by large corporations: pointing users to eat pizza with glue, providing false information about just about anything. There are a number of ethical issues yet to be resolved with usage of AI, from its intense usage of computing resources that adds up to electric grid demands. Through privacy and possible copyright violations of datasets that power the models. To an entire bag of other issues monitored by excellent resources such as AIAAIC repository. 

AI/Machine learning is an amazing field with many likely applications, and yet, its recent rise to fame is characterized by failures and issues in many implementations. Personally I often don't even see whether the application of AI is truthfully necessary, in many cases a human would do a more trustworthy and fast task of information gathering than a large machine learning model.

When we compare current state of "AI", does it reflect what Firefox stands for? Does it reflect Mozilla's principals?

Let's compare.

• Principle 2

  The internet is a global public resource that must remain open and accessible.

LLMs are known for being a black box. Depending on our definition of "open and accessible", LLMs can be a very free resource or a completely inaccessible black box of math.

• Principle 4

  Individuals’ security and privacy on the internet are fundamental and must not be treated as optional.

It's clear that LLMs pose a privacy risk to Internet users. LLMs pose risk in at least two ways - because the data they are trained on sometimes contains private information due to negligent training process. In this case users of a learned model can possibly access private information. The second risk is of course usage of 3rd party services that may use information to infringe on privacy of users. While Mozilla in blog assures that "we are committed to following the principles of user choice, agency, and privacy as we bring AI-powered enhancements to Firefox", it's unclear how supporting such services as "ChatGPT, Google Gemini, HuggingChat, and Le Chat Mistral" helps protect Firefox user privacy. Giving users choice should not compromise their safety and privacy.

• Principle 10

  Magnifying the public benefit aspects of the internet is an important goal, worthy of time, attention and commitment.

In my opinion in the process of designing AI functionalities on top of Firefox there was no evaluation on how those functionalities can benefit the public. There are a number of issues as mentioned above with the LLMs, they can be dangerous and work in detriment to users. Investing and supporting in technology of this type may lead to terrible consequences with little actual benefit.

In addition Mozilla claims:

• We are committed to an internet that elevates critical thinking, reasoned argument, shared knowledge, and verifiable facts.

I argue that AI models are the opposite to that. AI output is not verifiable. They are working against sharing knowledge by making seemingly accurate information that turn out to be false.

I ask Mozilla to reevaluate impact of AI considering all of those points. I ask on behalf of myself as well as many users that I see on Fediverse being greatly worried and frustrated with AI changes added on top of Firefox. There is certainly a lot of potential greatness that could be done with AI, but those steps must be taken responsibly.

---

Not just that, AI training actively incentivizes not just web scraping which we've already seen, but also harvesting data which is ostensibly private such as on "cloud" services like drop box, and potentially, call me paranoid, directly off your file system. Microsoft and OpenAI already have ties, I'm not saying it's currently happening, but OpenAI and every other AI company already have a "do first, ask for permission second" stand with all data harvesting.

If you want an example of precedent, Adobe are apparently using data on Adobe's cloud service for AI training.

[narazamsa]

If its on device only and no data collection is going to happen. Also only complete opensource models. Then yes that would be good to have.