Is #mastodon becoming an echo chamber?
-
@skyfaller@jawns.club I don't think that using LLMs is unacceptable. I don't buy the "built on stolen property" argument for this, every developer ever has searched for how to do something in stack overflow etc, I don't know why LLMs shouldn't be trained the same way in order to automate work for us. I get the climate impact argument, but again, flying with airplanes has much more of an impact, but it is normalized by now, practically nobody will tell you it's unacceptable to visit a foreign country by plane. @evan@cosocial.ca @carnage4life@mas.to @scottjenson@social.coop
@panos @evan @carnage4life @scottjenson This is what I'm talking about. You're both minimizing/denying harms and saying they don't matter. This is one of the biggest problems with LLMs, they turn people into apologists for the fossil fuel industry because they don't want to think they're helping destroy the world.
*If* flying is more harmful, that's no excuse. There's always something more harmful until you reach the top, and then the excuse will be it's too important or too difficult to stop.
-
@vfrmedia @mattwilcox
But even in these last two replies, you're perfectly making my point for me. Do I want AI Grifters? Of course not! But should people be allowed to discuss possible uses of AI? Absolutely! The person I quoted is NOT a grifter but due to our culture, he is seen as one and gleefully shoo'ed away.That is the slippery slope I'm worried about. We can likely agree on many bad actors that should not be here. What we can't seem to agree on is "what is honest debate"
@scottjenson @vfrmedia I don’t agree with the assertion people are being shooed away. Away from who? This isn’t one space.
But if Mastodon feel thy are, then make the tooling to solve that. And yeah; it’s the same tooling as black people wanted and it’s damning it takes rich white guys complaining to be considered as a problem worth fretting over and maybe tackling now.
Mastodon isn’t a monoculture. Mastodon is a tech stack.
-
@scottjenson @carnage4life Facebook, LinkedIn, and X don't allow this kind of API access. No one can keep us from building it here, though.
@scottjenson @carnage4life I think AI skeptics raise some very important questions. I love reading posts here -- especially when they skewer the conventional wisdom in Silicon Valley about productivity gains from AI. And the threat of further power concentration in big tech is very real.
-
@carnage4life @evan Strongly agree. The current AI companies have done much to be criticized but the tech itself, especially the open source and local versions (which this community should love) is actually a positive force here. We need to have discussions to understand the differences.
@scottjenson @carnage4life @evan I think step 1 is not calling all of it AI, an extremely broad vague term makes it very hard to have a nuanced discussion. If Mozilla uses a local ML model to detect which field on the page is which type to autofill better, that's very different from a remote LLM chat bot.
-
@evan @scottjenson @MozillaAI Mastodon as a community is quite hostile to AI and anything that isn’t a criticism of AI is viewed with skepticism at best and typically with hostility as the default.
It’s unfortunate because, as in your Mozilla example, there is still time to shape how AI is used in our industry. It’s better to engage and try to influence it versus stick your head in the sand and have the change thrusted upon you.
@carnage4life @evan @scottjenson @MozillaAI I’d probably be less hostile toward a harmful technology if it weren’t costing me and my friends so much of what we built, just so those who have disproportionately benefitted from our labor could take more short-term profits.
Nobody should be expected to apologize for standing up for themselves, their friends and colleagues, and what they’ve built together that is being poisoned.
-
@evan @cratermoon Exactly, even asking that question "What value do they bring" is kind of scary.
It's a fine line, I get it. We draw the line at nazis and scammers but let's not cross the line into "intellectual purity" tests
@scottjenson @evan The AI Boosters have used their money and influence to keep their ideas and views on blast from the highest levels. It's almost impossible *not* to hear their claims. What value does it bring to the FediVerse to have those claims repeated here?
-
@scottjenson @carnage4life I think AI skeptics raise some very important questions. I love reading posts here -- especially when they skewer the conventional wisdom in Silicon Valley about productivity gains from AI. And the threat of further power concentration in big tech is very real.
@scottjenson @carnage4life so, what can we do about it? One thing is just being brave enough to talk honestly about how AI affects your life and your work.
Another is calling out bad behaviour. If someone you know is yelling at a stranger to die in a fire because they used Claude Code, maybe give them some private feedback that it's not cool.
-
@scottjenson @Gargron “AI people” are not a protected class. It seems much more important that we focus on being a welcoming and inclusive platform for protected classes, particularly actual marginalized communities.
@scottjenson @Gargron it may be that “AI people” are sometimes not treated nicely or fairly. And we shouldn’t tolerate abuse. But I also really value that we rarely see AI hype which is often misinformation and disinformation that pollutes other platforms. That’s not just a personal preference but super important for a healthy media environment.
-
@scottjenson @carnage4life so, what can we do about it? One thing is just being brave enough to talk honestly about how AI affects your life and your work.
Another is calling out bad behaviour. If someone you know is yelling at a stranger to die in a fire because they used Claude Code, maybe give them some private feedback that it's not cool.
Lastly, and carefully, maybe we should put some technological speed bumps in the way of random abuse of strangers. Mastodon experimented with an AYS pop-up when replying to a stranger. I don't know what the results of that experiment were, or if any others are planned.
Improving the quality of conversations on Mastodon
In our most recent Mastodon for Android release, we’re testing a new feature aimed to curb unneccesary negativity that comes from being on the Internet.
Mastodon Blog (blog.joinmastodon.org)
-
@carnage4life @evan Strongly agree. The current AI companies have done much to be criticized but the tech itself, especially the open source and local versions (which this community should love) is actually a positive force here. We need to have discussions to understand the differences.
@scottjenson @carnage4life @evan There are some valid arguments to be against our current LLM-dominated AI landscape (and also some to be optimistic), The current debate is mostly "AI haters" vs "AI bros", which is a pity. It's extremely exhausting to constantly see these black and white takes.
Personally I want to be optimistic about this tech, but I'm not confident enough to see it as more than a fancy autocomplete at the moment, especially with the slew of security nightmares we've seen regarding agentic tools such as OpenClaw recently.
-
@osma The number don't lie, active users on Mastodon are going DOWN not up, we're hemorrhaging users, most likely because we're chasing many of them away with these 'purity tests'. I certainly am not experiencing this diversity you are referring to.
That's the point. Unmoderated servers see a completely different view, while the mainstream is heavily policed - and the majority of the still active seem to like it that way.
@scottjenson -
@scottjenson I’m not interested in following any “AI people”. That doesn’t make it an echo chamber. We don’t need equal amounts of people who love puppies and want to kill puppies, not everything needs to be equally represented.
@Gargron @scottjenson what about some light *kicking* of puppies?
-
@evan @cratermoon Exactly, even asking that question "What value do they bring" is kind of scary.
It's a fine line, I get it. We draw the line at nazis and scammers but let's not cross the line into "intellectual purity" tests
@scottjenson @evan @cratermoon
I feel like missing context. Let's layout it first:
- Mastodon is a technology for federated instances
- Admins make descisions about what is and is not on their instances. By virtue of control, admins can also enforce their own rules, on their own realms.
- It is federated, without central dictatum from whoever/whatever.What is the issue?
Are you talking about Mastodon org's policies?
Are you talking about some tech features that shift control? -
@scottjenson @carnage4life always has been, to be honest. Niche platform means you need to be fed up with mainstream enough to leave your friends, high barrier to entry and general decentralization struggles cause mostly technical folks to be here. Not surprised, that the opinions here are quite homogenous
@Igigog @scottjenson @carnage4life They are not homogenous, they are just louder than other voices. And they often sound like talking about a cult.
-
@scottjenson @Gargron it may be that “AI people” are sometimes not treated nicely or fairly. And we shouldn’t tolerate abuse. But I also really value that we rarely see AI hype which is often misinformation and disinformation that pollutes other platforms. That’s not just a personal preference but super important for a healthy media environment.
@scottjenson @Gargron Of course AI is a pretty vague concept and it’s not all hype and BS and I think your point holds for this. That said imploring people to be nice rarely works. It’s like driver safety education campaigns rarely do anything to improve safety. Systematic improvements for making this better for marginalized groups will also help non-hype AI people
-
@Igigog @scottjenson @carnage4life They are not homogenous, they are just louder than other voices. And they often sound like talking about a cult.
@Igigog @scottjenson @carnage4life (If we hadn't Bridgyfed and Flipboard, I would have left a long time ago.)
-
@scottjenson @vfrmedia I don’t agree with the assertion people are being shooed away. Away from who? This isn’t one space.
But if Mastodon feel thy are, then make the tooling to solve that. And yeah; it’s the same tooling as black people wanted and it’s damning it takes rich white guys complaining to be considered as a problem worth fretting over and maybe tackling now.
Mastodon isn’t a monoculture. Mastodon is a tech stack.
not getting "sufficient engagement" on toots isn't the same as being "shooed away".
I wouldn't consider the person being quoted as a grifter and TBH I agree with a lot of what he is saying, but until recently I thought he was some kind of marketing person who never really replied here and was just looking for a "one way platform" (like many who want to go viral/popular and promote a "personal brand").
Also what might seem "lukewarm" about politics can also be simply because everyone broadly agrees with you, and many don't want the "battlefield of ideas" culture that is prevalent on many other social networks..
-
@scottjenson @evan @cratermoon
I feel like missing context. Let's layout it first:
- Mastodon is a technology for federated instances
- Admins make descisions about what is and is not on their instances. By virtue of control, admins can also enforce their own rules, on their own realms.
- It is federated, without central dictatum from whoever/whatever.What is the issue?
Are you talking about Mastodon org's policies?
Are you talking about some tech features that shift control?@scottjenson @evan @cratermoon
On a human side:
- Media that doesn't kill my attention span will always be slower.
- Worthy ideas, take time to absorb.
- Look for pools with high signal to noise ratio. Especially for signals that stand the test of time.vs
Someone's engagement numbers are lower.
There is no floods of messages that ain't properly read. -
@evan @scottjenson @MozillaAI Mastodon as a community is quite hostile to AI and anything that isn’t a criticism of AI is viewed with skepticism at best and typically with hostility as the default.
It’s unfortunate because, as in your Mozilla example, there is still time to shape how AI is used in our industry. It’s better to engage and try to influence it versus stick your head in the sand and have the change thrusted upon you.
@carnage4life@mas.to @evan@cosocial.ca @scottjenson@social.coop @MozillaAI@mastodon.social I mean, Mozilla is doing exactly the same as other corpos, except in their case they just privacy-wash it
-
@scottjenson @Gargron There is nothing, absolutely nothing, stopping anyone from creating and cultivating an AI community on Mastodon. Start a server. Knock yourself out.
But expecting to *farm acceptance* from a group of people, one which most members vastly dislike AI, is quite the hubris.
But sure, the community at large is the problem.
Clean up your kitchen and maybe folks will join you for a meal.
@FeloniousPunk There ought to be an AI instance which all the bros and boosters can go to, and then I can block it in one go. That would be ultra-efficient.
Unfortunately the real fedi is a lot more messy.
