Petri Kajander

A room from Airchat's 2023 version, recorded as a video. The room's name is the title; the transcript below follows the video, and every time in it plays the video from that point.

Play video
Transcriptcaptions · 5:52

0:00 To new technologies like LLMs are also helping. So if you have 1 million token or even more or even 100k works, we can start to check whether people who are up for elections have actually done anything they have promised in the past. What is their track record? And these become like instant things. So verify, don't trust becomes the norm. What do you do when everything is fake? There's millions or trillions of infopecies around the world coming real time in. Nothing you can trust. Humanly, it's not possible to go through them. How do you get your information? Well, you get a sentiment, you get a trend. It's not truth. It has nothing to do with false or fake. But you get a sentiment what's happening. And we need to use some AI tools to get an aggregate view or some sense and trending what's happening in the world.

0:48 Interesting times. My hot take, news media will be dead. Or it doesn't die like anything from the past. We can still find it as a relic. Curious, peculiar, tiny, customized, luxury market, entertainment thing. For those who just missed the good old days. But for the rest, we only need the raw information sources, which will be then gathered, collected, aggregated, and served to us by AI and technical means. So no more journaling, no more reporting, no more these fakes or truths and whatnots. Times have changed. I think academia is just too slow. And it doesn't work in this environment. Stuff is happening so fast that you actually have to be in the middle of it. And you have to execute and you have to do things yourself.

1:38 And I think this has always been the way. David Deutsch, Einstein, they were not exactly in the center of the academia when they were doing this stuff. The other thing is that deep knowledge, just specialized, really deep but narrow knowledge, is also the specialization. It's something which is making you in trouble these days as well. So in a sense, the generalist may work better if you actually have the hands-on knowledge. So you can actually apply and use it to the new fields and do that. But that's where the shallowness comes in. If you're just superficially hopping from the wave of the next thing, and maybe you have a deep knowledge of something, but it's not applicable to the next one, or it's using different kind of knowledge space, then you're in trouble. And I see that quite a lot nowadays.

2:26 In some sense, it's like Nokia building excellent hardware, but you're already entering into the software world, and that's where they're not good at. So the meta is in a game where you need hardware, you need your customized chips, and not even that is good enough because you need the deep integration with the things only Apple can at this point provide. Even Google is struggling because they don't have the whole ecosystem and the value chain in a way Apple can integrate. Is this more like a UI paradox? So the data is in silos. Yeah, it's TikTok has already been taking over Google search, but the data is in TikTok, it's in YouTube. That's just one example or in Reddit. But is it more like that nobody goes to websites anymore because it's just more convenient to ask with your own voice and there's some kind of Siri answering.

3:19 So it's voice interface, it's some kind of a chat interface. So the data, the data layer kind of disappears. You just wish and it happens. And this changes the whole nature of internet. Is it becoming like a huge silo of data with APIs and the visible layer starts to vanish on the background, which results in a different way of accessing, also more difficult for the open data, open source data, unless you start on your own data. So it becomes completely decentralized, atomized with its ownership that nobody except you has the permission access and you can give the read access, write access to whoever you like.

4:10 And that way maybe the power is taken away from the big ones, but yeah, just a thought. TikTok might be the third one who has enough videos and they are also educational ones. So it's not just dance moves and silly memes. They are clearly having the muscle and the power to do that. Obviously dominant in China, but also I would assume that the English one has enough stuff to do something about it. And they are notorious of getting your metadata and stealing your clipboard even. Threads is good for commercial use. It's good for those who are using the advertising model. So consumer brands, but also news agents, agencies, media outlets putting out their news pieces. Social media managers for brands are already using Instagram.

5:02 So Threads is the new IT. And this is where if you look, they're not so comfortable in Twitter. Many of the brands, some of them are obviously, but most of them are not. And Mastodon is not really a place for brands. So there isn't really any other place to go, but now they have more text. They can express themselves. So I would say that Meta becomes, Threads becomes the commercial space, the middle ground. It's also probably the new Facebook. If you went in the early days of Facebook to talk and chat, and it was a friendly, easy threshold place to express your ideas. This is now the opportunity for these people who are, don't like or don't know what Twitter is, or just are not there or are not good at that. So I would say that Threads is a good place to go.

5:52 But they want to do something more than just put up pretty pictures, which was not their thing. So this is the new Facebook for those people. So I think there's a place for these type of people, and it's probably quite a large amount of people.