EP 394: AI News That Matters - November 4th, 2024
37:20 Share

This is the everyday A I show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business and everyday life.

ChatGPT s new search tool is out, but that was actually kind of scooped by google. Mas lama model is reportedly being used by the chinese military. Microsoft and google are experiencing A I delays anthropic isn't they keep shipping and ChatGPT won't be shipping GPT anytime soon.

Yeah of what going on in A I news this week, as always, and we're going to get you caught up right now what's going on your my name is john Wilson and i'm the host of everyday AI. welcome. Thank you for tuning in. If you're listening on the podcast, appreciate you. As always, make sure to check about your show notes for more on all of today's stories that we're going to be recap ing lifestream audience, thank you.

It's great up to see i'm actually in a different country in australia right now on the anniversary with my wife uh so you know who knew shot out to sydney apparently out check in our our podcast at sydney is the number one city all time as a thanks uh for uh kind of keeping keeping with today's audio and video issues. Not going to be the best a try and very hard but coming to live ish in uh, the hotel room here. So h sorry, there's a little bit of eco but and not for that, let just get into what actually matters and going over the AI news.

And like I said, if you haven't already, please go to your everyday aides outcome sound up for the free day in newsletter. We will be recapped all of these stories. All right, let's get straight into IT.

That first story, ChatGPT has introduced its in web search capability for users. So ChatGPT has just rolled out in upgraded search feature that allows users to receive timely answers with links to relevant web sources, significantly improve improving the user experience. So the new search functionality is available to ChatGPT plus and teams users with plans to extend uh access to free tears in the coming months.

So the feature enables users to ask questions conversationally and receive responses that may include information source from the web, making IT easier to obtain accurate, irrelevant answers and then follow up questions will be considered in the context of the entire chat, allowing for deeper engagement and more Taylor red responses. So this is essentially what we heard of a while back when we heard of this search GPT. Uh, so when open eye released that they did say search GPT was a prototype and what was just released does not look exactly like this, but IT does look like a much improved version of open a order browse with being a search.

So the new chat P T search, i've used IT i'm sure many of you have as well. If you haven't used IT, IT really just kind of looks and feels a little more like google, right? So that's what I think open the eyes is really making a huge play for because here's what people don't understand.

Uh, right now you do not even need an account uh, to use ChatGPT. You don't need to be logged in. You can literally just go to ChatGPT dot com and start using IT so you can use this new ChatGPT search that way.

Again, right now, you have to have an account and IT needs to be a paid account. But I think this is where OpenAI is actually headed right now. I don't think they're trying to compete with perplexity necessarily.

I think they're actually trying to compete with google. I I think is a smart play because um the new chat g GPT search is really, really good. Uh, so feedback from global publishers so far has been instrumental in shipping this feature, emphasizing the importance of maintaining journalism core values and integrity.

So that's another big piece here. OpenAI has a lot of deals with publishers that is allowing them to source this material um more quickly. So yeah, this is blazing fast.

This is faster than perplexity and IT is even faster than google A I search. So uh it's very impressive to see um what OpenAI has done so far with ChatGPT search. And i'm sure we'll be talking about this a lot more.

alright. Well, they actually kind of got scooped by this stop because google literally right before this, they launched their real time search for gemini A I, setting the stage for some intense competition in the A I power search. So google has introduced real time search capabilities, but for its gi AI platform, not for its gami chat bott. So the new feature is called grounding with google search, and IT enables developers to enhance their AI applications with current information source directly from google search. So like I said, um this is not the same thing as ChatGPT search or complexity.

This is for developers who are already may be using gemini, but to get real time access to google search, which I think is huge, right? But again, if you're just using A A chat pots as a front end user, or if your company maybe uses the german I A P I, uh, but you don't need real searched, this may not impact you. However, if you are using the gi API or maybe you want to learn IT and you also need real time access, uh, to google.

This is huge. Uh, so you know you have to give you kind of tip your hat to google. I've been saying this for a very long time.

I am not impressed sed at all with their german I chat, but they're back. And tools for developers. fantastic. Uh, so this new grounding a with google search pretty cool. So the service right now is Price at thirty five dollars per one thousand query.

So kind of Prices, right, but reflecting the high computing cost associated with this real time search functionality um on top of the gami API. So google google dynamic retribution system scores queries based on their relevance to current events, with scores ranging from zero to one to optimize costs in response time. So yet there's a little bit of handling on the front end uh there.

So this integration comes at a critical moment as google reported forty nine point four billion dollars in search search advertising revenue ah in the quarter three when they just reported uh their quarterly h earnings. While also facing increasing competition from AI driven initiatives. And like we said, open the eyes.

Recent launch of ChatGPT search offers a consumer focus approach. Where is this new crowding with google search is really offering a similar capability, but just for developers. So both companies have secured licensing agreements with major news organizations, although the details of these deals remained undisclosed.

And it's not with all the publishers, right? Here's the the big you know elephant in the room that mostly people aren't talking about, right? All these big large language model makers essentially just scrape all website content, not just those that they have licensing deals with.

So IT does seem like some in some unofficial al testing early on that they seem to prefer or to serve answers from those publishers that actually have, uh, you know, the results. So know where you can test. This is very unofficial, al. You can do a Normal like google search as an example and see you know which of these partnership uh, players that maybe have a result of, uh you know the first page of google if you're searching for something and then you can do that same h search.

You know if you are using the grounding with google search or similarly, you do something like that with ChatGPT search and you will see that open the eye and google are seemingly um you favoring these companies that they have licensing deals with, but that doesn't cover everything, right? So don't worry if if if you're publisher out there, you know your results are still in be coming up on a ChatGPT search in this new grounding with google search. This is Jordan, the host of everyday AI.

I've spent more than a thousand hours inside ChatGPT, and i'm sharing all of my secrets in our free prime prompt publish ChatGPT course that's only available to loyal listeners like you take IT from ban, a company founder who took the ppp course. Your prime prompt in polish class was absolutely fantastic. If you're new to A I like me, a leap frogs, you're learning immensely.

If you have been playing around with IT for a little while, I can imagine how I would help you just elevate. You're learning that much faster. So I encourage you all to give IT a try and check IT out everyone's prompting.

Wrong in the ppp course fixes that if you want access, go to pog ppp dot com again, that's pod P P P dot com set for the free course and start putting ChatGPT to work for you. Our our next piece of A I news, speaking of the big large language model makers, claude just shipped a lot, right? So h claud launched their dedicated mac APP.

So they've introduced a dedicated mac APP for claud in dancing accessibility for users on both mac and windows platforms. So the claude APP is currently in beta and available for download to all users, whether they are on free or paid plans. So we got ta love that from anthropic.

So the APP right now provides easy access to claudes latest model, which is clad three point five unit new yeah I know you can just I call IT three point six. A lot of people call three point six. Uh, but this also um is not the only update because there's also dictation with in the APP ah and also two other things from anthropic.

So uh they also released the new P D F processing system that employs a three days methodology which is text extraction, visual processing and integrated analysis allowing for a more comprehensive understanding of documents. So what that means yet, let me just recap that. So we have now the new number one, we have the new desktop APP.

Uh, number two, there is dictation now so you can speak to claud and you know instead of having to type to IT and number three, IT has this new kind of three pronged PDF support. And this is huge because by default, most large language models do not handle kind of, uh, i'll say, multi moto P D S, right, even though they're not multi model technically right. But essentially a most large, large models don't do well.

If the PDF that you upload and you're asking IT a question, or you're asking IT to read something back or somewhere ize something, most do not do a good job with images, graphics, inc. tra. So this is really what this update from anthropic is all about.

A the the PDF update, uh, is this new kind of three face methodology, text extraction, which is not new, but the big thing here is the visual processing and the integrated analysis. So uh, I did do a quick uh kind of little tip trick tutorial on how to actually do this in ChatGPT because by default, ChatGPT doesn't do this very well. You do have that you know know your way round some basic prob engineering.

So love to see this update from anthropic and hate live stream audience. Let me know if you want to see maybe a tutorial on these new a PDF features from anthropic. Ah we got so much news so well. Anthropic was announcing updates and just shipping. Google was announcing delays are so google has delayed its project astra A I launch until at least twenty twenty five.

So google's ambitious project astra, which was aimed at developing A I applications for a real time understanding, will not be launched until at least twenty twenty five, as revealed by google, CEO and dark patched during the company's order three earnings hole. So project astra was showcase and in a demo at google's I O developer conference in may of this year, highlights potential to enhance smartphone capabilities such as recognizing surroundings and and in questions. So essentially what this project astra was IT was what ChatGPT alive, uh, kind of voice assistance will be or could be once IT gets video capabilities right?

So uh, what they showed in the demo was, you know, someone kind of walking around google office asking questions about things on the desk can know kind of use computer vision and see in real time. It's pretty interesting here that google had yet a another delay. Uh, you will also remember about a year ago, google infamous ously kind of showcase some of these features right when I first launched its gi kind of chat out but then they kind of later said, ah yeah we I guess everyone else kind of said like this isn't really true.

This is not how IT works ah so google did have a live working in demo of project aster which again IT can see in real time with video and then IT is a voice assistant that you can talk to so IT did appear to be a working demo uh in may at the A I O conference. Yet we might not even get IT in twenty twenty five yes, IT says twenty twenty five at the earliest. So IT might not even be until l twenty and twenty six.

So uh this is another page of the google playbook with uh you know shiny demo ah you know lots of marketing and then you know you might get IT in three to six months or in this case even longer. Uh so pretty pretty shocking that we might not even get IT uh, in twenty twenty five. The initial reports suggested that a consumer focused agent experience was set to launch in december twenty twenty four, and IT was capable of handling tasks like purchasing products and booking flights.

However, we don't know if this report still holds true. We talked about this last week on R A I news that matter segment, so we weren't sure that is more similar to claudes uh, computer use, right? So more of a of agent that you run on your computer or if this was supposed to be project astra.

But now at least according to google CEO. So from the source himself, we are not going to beginning project astra. So pretty interesting. There are a lot of delays and there not the only one because microsoft also just delayed its I don't even know if we can call IT a popular feature because IT has not been released yet, but microsoft has once again delayed its recall feature for copilot plus pcs until december are. So at least we have a time line on this one.

So microsoft announced another way in the rollout of its recall feature for copilot plus pcs, which is now set to be previewed a with windows insiders by december. So the recall feature essentially IT is like history for your browse ser, but for everything, right? So anything that you do within windows programs, you can kind of remember IT and then you can talk to IT like you would a large language model.

And there's when I was first announced, I think there was both a lot of optimism but also a lot of skettles t like scepticism where right people are like old. This is a privacy nightmare. And I think microsoft did a good job in addressing those concerns ah.

But again, we are seeing more delays. So uh, the recall feature was originally scheduled to be tested with windows insiders in october but has, like we said, just been postponed till at least december for further refinement. So like I said, recall IT captures screen shots of user activity on copilot plus pcs, and IT aims to enhances productivity by allowing users to search and retrieve their past actions.

Ah so the features development has been complicated by security issues leading to an overhang that included making recall and opt in experience and ensuring a complete uninstall option for users. So yes, whether this was a kind of feature that was added or a security hole that microsoft did plug. But you know now IT is an opt in feature.

Where's before IT was presumed to be a kind of a feature that was on for everyone and then you had to turn that off if you did not want IT. Ah so microsoft did clarify this a couple a month ago and say IT is in opt in experience and um also IT does give the option for a uh complete uninstall for users. There's also been a little more confusion with some mistakenly believing recall was being installed on all pcs running windows eleven, but that is not the case.

IT is just for those copilot plus pcs. So these are for the newer uh copilot plus pcs h that have the know very powerful CPU G P U and also the NPU, which uh allows you to run kind of this edge A I or on device A I. Alright, so yeah, microsoft made a ton of headlines that we just talked about.

Two of them are also get they just had a some huge um updates with github coal pilots. But here is one that i'm not maybe a few enough. So according to reports from a tech rate or microsoft is reportedly considering a rebranding of its AI features under the name windows intelligence, potentially of lining its street teg with apple's branding choices when apple decided to call their A I apple intelligence.

So again, this is just according to reports from tech radar, and nothing official here yet, but references from an internal Operating file suggests that A I features in windows of eleven may soon be grouped under a windows intelligence name, indicating a potential strategic rebranding effort. So the name windows intelligence has been used by microsoft in the context of cyber security, but this new application could signify a broader A I integration by adopting a named similar to apples apple intelligence. Microsoft appears, according to reports, to be aiming to create a strong association between its Operating system and A I capabilities in the mind of consumers.

Uh, let me just i'm going to go in and insert my opinion right. Normally I don't do this too much on our a weekly A I news that matters shows. I don't think this is needed. I don't I think the apple intelligence moniker is kind of dumb if i'm being honest.

Uh, you know, I don't think you because apple was so late to the party, you years years after microsoft, years after google, years after chat, P, T and apple comes in and just, you know they say, oh, this is apple intelligence right? And for the most part, yes, IT does do some things are on device and IT does so okay, nothing great, right? Uh the the best feature of apple intelligence so far is its integration with ChatGPT.

So I I don't think that microsoft needs to rebrand uh its A I offerings under a windows intelligence. Uh, not a huge fan. I personally like the word copilot, right? That's where it's it's it's branded right now. I think that so many people out there are are using this copilot terminology, uh, to talk about or to describe an AI that you can work along with. Um however, microsoft so I get that microsoft kind of in a weird position with this right, because everyone's using copilot to reference in A I right.

But I would say the majority of people when you talk about A I and copilot, you're thinking microsoft, right? So it's almost like a clean x right? You don't tell someone pay, pass me tissue past me unnamed, what what you're right.

You say clean ex, what? There's a clean ex or not. And I think that microsoft has gotten to that point a with its copilot. So i'm not sure really why they would want to be moving away again.

This is no they probably have dozens of of people working on this project who know way much, who know way more than I do, but not not a huge friend, but the rebranding of copilot, um this isn't anything new, right? So we saw a complete rebrand if you're using co pilot on the web. So uh, if you're using the free version of copilot or the paid version, so not the.

Biz chat, but otherwise you didn't see a completely new a version of microsoft copilot, right? Saw rebranded completely new interface with the new voices. So yet we might see microsoft continue in this direction.

All right. AI agents at linton. Yeah, maybe so. Lindon has just launched in A I hiring assistant that they're hoping will transform the recruiting process. So linton has introduced its first AI hiring assistant, aiming to revolutionize the recruiting landscape by automating up to eighty percent of traditional recruiting test, which is wild.

So the new AI hiring assistant is designed to streamline various recruiting task, from creating job descriptions to conducting interviews, spring up valuable time for recruiters, so the tool bills on lindon's previous features, enhancing the capabilities of its generated A I to offer personalized suggestions through conversational memory, so recruiters can input job descriptions, notes and a feedback, ensuring the ais output of lines with their unique requirements. So concerns remain, though, that the increase reliance on A I could lead to bias in a lack of personal touch in canada interactions, as seen in similar instances with other companies like amazon. However, that allows recruiters to selectively automate tasks, enabling them to maintain the human elements essential in recruiting.

So the other thing I mean this is this is great also, you know pretty expected they didn't has been a pretty heavy uh, on A I given IT is uh owned by microsoft, right? However, don't fully understand this one either because unless they're going to end up charging a ton for this, right, if you're going to automate eighty percent of traditional recruiting task, isn't this also going to cut down their revenue a lot? I mean, i'm sure they've bought this through.

But think, think of this, let's say you have a team, let's say you're big recruiting agency, right? And you have a team of one hundred recruiters, right? Is this only going to automate eighty percent of all of those recruiters jobs? Or might you only need two of them?

In theory, if this works out so um I like to push here by link down because I think so many of these task in recruiting, right, I i've spent so much time recruiting people to small you know small businesses that I own as well as companies that i've worked with the past so I understand that a lot of this is mundane and IT should and can be automated but say that you can automate up to eighty percent um I mean you have to think like, okay, is this either going to be terribly expensive right once it's kind of fully roll out or are just you know why you have to if you have a team of a hundred recruiters, maybe is your team going to be fifty or maybe is that only going going to be twenty, right? So it's kind of weird, right, that laden is hoping to automate, uh, a big part of its markey software, right? A linton premium, linton sales navigator, sales navigator linton recruit, right? So yeah, interesting move.

So we'll have to see how this one pans out once IT fully comes to market. alright. Uh, our next news story ah for this week, china is leveraging metas lama model for military AI development, raising security concerns.

So recent reports from reuters indicated the chinese that chinese researchers linked to the people's liberation army have adapted metas alama model for military applications, highlighting the ongoing global competition in artificial intelligence. So a june paper from six researchers associated with the academy, the academy of military science detail. The creation of chat B I T or chat bit.

I'm not sure which one of is in A I I think it's actually chat B I T in the AI two are based on an early version of meta lama model, specifically tailored for military intelligence and Operational decision making. So chat bit reportedly outperformed other AI models, achieving capabilities comparable to open the eyes ChatGPT four. Although the researchers provided limited details on its performance, metrix and Operational status, meet has stated that any military use of its open weights models by the P, L A is unauthorized in contradicts its acceptable use policy, which prohibits applications in military warfare and espana context.

So despite these restrictions, metal acknowledges the chAllenges of enforcing its policies due to the public availability of its models, which has LED to concerns about the potential mix of open source A I technology. Yeah, so I think we've kind of known that this has been going on for quite some time, but not specifically right. Uh, I think anyone that that reads a lot of A I news or like myself, that is to talk to very far people, right? We all know that these open source models are being used for bad purposes.

Sometimes there's exact, sometimes there's not. right? So here, this exclusive report from reuters, a little scary, right, that china has been using this for, uh, these open models for military purposes.

So, uh, this does raise some concerns on ways that you can Better track how these models are being used and for what purposes. And I mean, we might see in the same way that the U. S.

Government has recently enforced some restrictions on company's exporting GPU. As an example, we might see some similar restrictions on uh, you know exporting models. And I know that may not make sense now, right.

But think back in the day, you know downloading music right and uh you know you can ban ips or something like that. So uh, you know I wouldn't be surprised if we see a move from the U. S.

Government sometime soon because as metas models specifically get Better and Better and Better, uh, you know these cloth in proprietary models, you have so much more control over knowing who's using them and who's not right. Because in theory, the companies know that. But at a certain point, right, when these open source models, they get forked, you know, they get fine tune and and we uploaded, right? So that is the nature of an open and open source. Open weights.

I know that is not truly open source, but with an open weight model is IT just keeps being built upon, right? Is almost like a cup of water, right? If if you keep pouring IT in ten different glasses, pretty soon you're going to forget where IT came from, which is problematic, right? So that is the the downside in the ugly side of these models that are very open is IT can end up in the bad, like bad hands has that you don't want IT into such as U.

S. Kind of enemy forces in these chinese military Operations, all right. And our last A I news story of the day didn't see this one coming, but OpenAI CEO sam altman, in a reddit, asked me anything session discussed well, just about everything.

But one of the biggest thing here was discussing the GPT five time line in some future A I releases. So in a recent reddit, ask me anything post OpenAIco an d al tman ad dress th e an ticipated re lease of GP T fi ve an d al so pr ovided in sights on th e co mpany's on going st ruggles wi th co mpute ca pacity, which key said is affecting product time minds. So alt men clarify that there is no imminent released plan for GPT five, stating we have some very good releases coming later this year.

But he emphasized that none of them will barathrum GPT five name. So you have a lot of speculation about what there's that mean. You know people are saying, oh could still be GPT five but I could be this new uh ryan model or could be a GPT four five but sam Allen did essentially say there won't be a GPT five uh, this year, which is something that um I had to be saying this saying this for forever.

I said there will not be a GPT five in twenty twenty four. I said that would probably be first quarter of twenty twenty five if we do actually get IT. Ah so the ongoing chAllenges are in compute capacity are also is a significant factor in delaying the rollout of these new AI models, altman said.

And he also noted that as models become more complex, OpenAI faces some tough decisions about resource allocation. right? They have all these other products that also require a lot of resources.

But recent reports, which we talked about in the news letter last week, suggest that OpenAI has been working with brought com to develop a specialized A I chip to enhance its computational power, although this new technology may not be available until twenty twenty six, if it's available at all. So some other tidbits from of this ask me anything session. So the advanced voice mode feature for ChatGPT was initially expected to include vision capabilities, but those plans have been postponed again due to compute limitations.

right? So same moment saying, hey, we don't have enough access to compute. Ah so the rushed demo of that feature was during a press event was intended to just draw attention away from google I O developer conference despite concerns about its readily ess.

So there will go again, you know having these working demos really just being used as a marketing against the other big company. And you here we go, it's it's I mean, who's actually shipping here. So also often mentioned that the image generation tool dolly uh is currently lacking any release tied lines for future versions.

And the video generation tool sora has also encountered some technical setbacks that hinder its competitive edge against other AI video systems like luma and runway. Uh so all men reiterated that improving the reasoning models right now. So the o one preview, uh the o one mini and eventually the o one four model are open A S top priority with several exciting features on the horizon.

All right. So yeah pretty, pretty interesting. I again, i'm on vacation, so I was not expecting same altman to do IT and tell us all, kind of redit ask me anything. So nothing. They're a terribly shocking aside from him saying that the top priority was seemingly working on the reasoning models and that a GPT five model would not be released this uh, this year where we did have previous reporting as recent as last week saying that a new model would potentially be landing in december after the U. S.

Election, right so uh we've always said here the election, uh tomorrow, uh, right we've always said that there will not be uh in by we I voice said that there will not be, uh you know a big claud saw IT release which there wasn't I said there won't be a google gami two point of which there wasn't I said there won't be GPT five there wasn't until after the election even though every other person out there know a year ago was saying, so we're going to get these models i'm like, no, we're not not until after the election and then know we kind of heard these december a time minds but IT looks like, uh, at least for GPT five, we are not going to beginning IT in twenty twenty four and the priority is the reasoning models right? So that was a lot you all let me just do A A very, very quick recap of the AI news that matters for this week, november forth. So ChatGPT introduced its enhanced web search capability.

So not exactly the search GPT, but IT does look like this is going to be. You know, kind of the replacement for that very capable, very impressive. Uh, next google, in response, or actually right before open a eye, did launch its real time at search for geri A I in the A, P, I, setting the stage for some intense competition in A I powered search.

Uh, speaking of launches, cloud also launched its dedicated mac APP. Its dictation features in a new three tiered PDF processing system inside a cloud. Google announced delayed for announced delays for its project astra and said IT might not launch until twenty twenty five at the earliest.

Speaking of delays, microsoft also announced another round of delays for its copilot plus pcs ability for the recall feature. But uh, IT looks like now we make IT that in december. Then we have microsoft a report from tech radar that said microsoft are work, may rebrand its AI features as windows intelligence.

I hope not, but could happen also. Linton launched its A I hiring assistant. Uh, china is reportedly leveraging metas lama model for military AI development.

And we got a whole bunch of OpenAI and ChatGPT news. When sam alton did, they ask me anything on redit, right? That was a lot as IT is every single week year.

I hope this was helpful. I'm sorry that that the audio and the video is not as good as that Normally is, but will be back, uh, next week for more. So thank you for tuning in.

And hey, shout out australia. I'm enjoying IT here so far. If if you're listened on the podcast, if you're still listening, if you're from australia, reach out.

I would love to just, you know give me a recommendation or two. So thank you all for tuning in. I hope see you all tomorrow and every day for more everyday A I thank so.

And that's a rap for today's edition of everyday A I. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. IT helps keep us going for a little more A I magic, visit your everyday A I dot com and sign up to our daily newsletters so you don't get left behind. Go breaks some barriers and we'll see you next time.