Both Microsoft and Apple are ditching their board seats at OpenAI...

open ey just announced a new bioscience partnership and at the same time Microsoft and Apple both Dro their openi seats their board seats as AI takes over make this your Mantra let the robots do the work subscribe to stay on top of AI news now Microsoft was already on the board of openi and there was speculation that Apple will be joining the board as well in fact in the last couple days they've announced that it was happening but it seems like they did a full 180 and are saying that they're both either leaving the board or not planning to take part of the board effective immediately there seems to be a lot of urgency because again Microsoft is uh basically sending them a letter saying right now right away and apple saying yeah we're not going in there nope and they're declining to comment as to why now what could be behind this move well there's some Anti-Trust worries both in the US and the EU they're kind of looking at that partnership between these big Monopoly companies and open Ai and they're getting a little bit worried now the official position by Microsoft at least was that the position that they had on open eyes board and provided insights into the board's activities without compromising its independence he was sort of an observer role looking into what opena was doing but not really interfering however since then they've witnessed significant progress from the newly formed board and they're confident in the company's Direction the role is no longer necessary now it looks like both the European Commission said in June it was exploring the possibility of an antitrust investigation same thing with the FDC here in the US Lisa KH has made a few appearances talking about the problems with these big Tech monopolies and how AI plays into it and it does seem like this move to not take these board positions open thei it's unlikely to resolve the agency's concerns but it does seem like it's going to make it harder for these antitrust Regulators are on both sides of the Atlantic us and EU to prove that Microsoft has you know un do influence over openai it's going to make it harder for them to prove otherwise in the meantime openai in Los Almos National Laboratory announce bioscience research partnership what's interesting to me here is that the Manhattan Project the project to discover the nuclear bomb did happened in Los Almos as well but we'll we'll get back to that in just a second because there's a few more announcements that are just happening right now the first one is that open eyes is rolling out its text to speech API to the playground so you're able to just kind of play around with it produce voices they have six different voices that you can use and kind of get your hands on it and test it out in the playground the open AI playground and interestingly it does look like you can download it which is kind of a game changer because you could use this basically for free I've used the text to speech with the I think they have six different voices used it for The Voice Assistant that I've showed on this channel it works great I think here the added benefit the added functionality is that you can actually just download whatever thing you have which almost kind of seems like it kind of put them on a competition course with 11 Labs of course 11 Labs that's kind of the whole point of it is you create voices that you type in whatever you want the AI voice says that thing and then you're able to download that audio file and use it wherever you want to so here's kind of what that looks like the text of speech appears here you're able to choose from six different voices allo is the default you're able to do text of speech one or text of speech one HD as well as looks like the playback how quickly they say it and whether you want it as an MP3 or a wave file or something similar let's try it out fly you fools and they have these Generations on the right hand side here and let's try a different voice echo get to the chopper get to the chopper and I can download that and have that file that is pretty cool I got to say neat in other news Mir Madi is saying that Sora is not quite ready for prime time yet take a listen speaking of Sora this week uh Ashton Kutcher told Eric Schmidt what an interesting par um I have a beta version of it and it's pretty amazing he also said the bar is going to go way up because why are you going to watch my movie when you could just watch your own movie he's basically saying people will be able to type out their movie in their head and AI will turn in a movie they can watch right now it's quite crude but it does it sound plausible to you and and when will Sora be ready for public uh release we don't have a timeline for public release for Sora yet uh what we're doing right now with Sora is we we've given access to red teamers and we've given access to some content creators to help us identify ways to make this robust we're doing a lot of work on safety front uh but also to figure out how do we actually bring this to the public in a way that's useful um that's not very straightforward right now it's really a technology um and this has been a pretty consistent process that we have followed with every new technology that we have developed we usually work with those that have uh like for example with Del we worked with creators initially and they helped us identify ways to create an interface that where they felt you know more empowered and they could do uh they could create more projects and make yeah basically you just want to extend the creativity of but let's come back to this open ey and Los Almos partnership in biosciences so the point of it is to understand how AI can be used safely by scientists in laboratory settings to advance bioscientific research here in the US they passed a safe AI law the development and use of artificial intelligence where they tasked the US Department of Defense National Labs to help evaluate the capability of these Frontier AI models the best of the best including their ability in various biological tasks creating potentially some bio viruses or anything that could be harmful so it could be viruses drugs compounds you name it it looks like one of the interesting things that they're testing is they're seeing how GPT 40 can assist humans with performing tasks in a physical lab setting through multimodal capabilities like vision and voice all right so didn't mention that this feature is yet unreleased but this is the real-time voice system that we're hoping to see sometime later this year but from playing around with building my own kind of little voice assistance I use the text to speech API that I've just shown in in the previous segment the assistant itself could be gemini or GPT 40 now more and more it's Claude 3.5 Sonet and you can have it be a voice assistant with no vision or you can hook it up to your webcam so it's able to see what your web camera is seeing and respond to questions about that or it can respond to screenshots of your computer so certainly having an assistant like that helping you as you work in a lab or office or anywhere else I could see the potential for that and of course the other big part is the bio risk threat evaluation to see is it possible these AI models create compounds or output any text that can help with the creation of some sort of a bio threat or can create any text that will help with the creation of anything that has any biothreat risk it's interesting that they're testing things that serve as a proxy for more complex tasks that pose a dual use concern they mentioned tasks like introducing foreign genetic material into a host organism maintaining and propagating cells in vitro and cell separation through centrifugation so this is very interesting because of course this could really upskill existing professionals phds as well as novices it could help them get the information they need upskill them walk them through certain steps remind them if they're missing anything answer questions and of course it can be used for nefarious tasks which is I think the thing that they're trying to prevent this is Lisa Khan the chair of the Federal Trade Commission and she's got a bone to pick with these big Tech companies I don't think she likes them very much but overall listening to her talk about AI the need for open source the need for decentralization so not having completely monopolized in the hands of the few big tech companies certainly a lot of the things that she's saying kind of make sense to me let me know what you think here's a clip and we as policy makers and as a society can help make decisions and choices that are going to steer these Technologies on a path that actually serves us rather than a model where a handful of companies are just extracting more and more from society from creators and people feel they don't have recourse and I would say on the antitrust side we are scrutinizing the entire stack from the chips to the compute and Cloud to the models to the apps trying to understand what are the kind of Key properties of each of these layers in layers where you are already seeing some bottlenecks or a handful of companies controlling that stack how do we make sure that's not being used in anti-competitive ways such that innovators and entrepreneurs are going to be locked out I went to Silicon Valley a few months ago and I heard a tremendous amount of excitement from a lot of Founders and startups especially about the potential of Open Source which has historically been a key driver of innovation but there was also some apprehension in part because there was a recognition that right now the raw material for a lot of these tools is in the hands of a very small number of companies and so who's even getting access to the models who's getting access to compute and on what terms all of that seems very opaque to a lot of the founders in startups and I think there's a concern that especially if you have vertical integration and conflicts of interests there could be self-dealing there could be discrimination there could be exclusion so that the big guys are just getting bigger at the expense of everybody else well is that the logical maybe inevitable outcome of this process that the big players who are big now become the big players in the AI space inevitably or is there a path that's where that's not the case I think this is where you need to look at each layer of the stack and so for example we see you know significant consolidation in the cloud computing space you basically have three major players there uh right now we see significant consolidation in the chips layer as well and we need to understand is there a path for injecting more competition into these layers um at the very least we need to make sure that these firms are not using any Monopoly power or Market power that they may have to entrench that dominance or to exclude competitors let me know what you think about that what is opene and Microsoft leaving very very rapidly very quickly the board of open AI what does that mean for the future of AI what does it mean that both both the EU and the US authorities are seemingly very interested in breaking up some of these what they perceive to be potential AI monopolies is this good bad let me know in the comments my name is Wes Roth and thank you for watching

Channel: Wes Roth

Views: 13,094

Keywords:

Id: gqWXgS-OYXY

Length: 11min 15sec (675 seconds)

Published: Wed Jul 10 2024