More Than 1/4 of Google's Code Now Generated by AI
14:56 Share

Today, on the daily brief, google reveals that more than a quarter of their new code is now generated by artificial intelligence. Before that, in the headlines, elan's X A I is apparently raising new funding at a forty billion dollar valuation.

Daily brief is a daily podcast and video about the most important news and discussions in A I to join the conversation fall the display in our showed tes welcome back to the AI daily brief headlines edition. All the daily AI knows you need in around five minutes. We kick off today with the latest funding rumors in the wake of open the eyes, massive one hundred and fifty seven billion dollar valuation round.

Elon musk X A, I is looking to raise billions as well at evaluation that, if not, the stratospheric one hundred and fifty seven billion is still very high. According to wall street journal reporting, the companies in talks to raise at a forty billion dollar evaluation. The news comes just five months after X A I S previous round, which rays six billion on a twenty four billion lar valuation.

Since then, the companies is completed construction of the colossal supercomputer at record speed and rolled out multiple new features, meaning that to some investors, that jump and evaluation is likely justified. The journal emphasize that the deal is still in its early stages, so is subject to change or abandonment. Mosque plans for the money are pretty clear.

On a video called tuesday into a conference in study a ababa, musk said, if you're training a frontier model, you need a massive amount of compute. On monday, he confirm plans to double the size of the glosses training cluster by adding an additional one hundred thousand gp use nica C A venki rights. Elan wants to true open a ee competitor and say what you will, but Ellen's companies rarely failed to raise the money.

They ask for the elan facts account, which I will let you decide what it's likely. Bias is pointed out that in video, C E O genco hong recently said, building a massive supercomputer factory in the short time that was done, that is superhuman. There's only one person in the world who could do that. Ellen and the X A I team did a singular, never been done before. Every R, D, S, ja zang rights underrated X, C, I and google accurately rated OpenAI anthropic meta.

And I actually think that the bigger point here is not so much the underrated or accurately related designation, but the fact that when IT comes to frontier models, if you are actually trying to compete in that space, if you are an investor who has a theis, that the value proposition of agi is going to be so large, that effectively any Prices worthy to invest in these plus mystery are pretty much the only way you can make that bet, which means that the Price is fairly in elastic. Now we had previously gotten reports that anthropic had also floated forty billion in its own funding talks. And my guesses is that if ilan is coming anywhere, you're close to that and tropics gone to use that fact to try to go for something even higher than forty billion, making the argument that they are proportionally farther ahead than the u.

On is who knows though, we will have to wait and see. Meanwhile, some interesting hardware news. Open eye will reportedly have their own custom designed AI chip ready to use by twenty twenty six.

According to reid's, the company has explored all options to diversify their chips supply and reduce costs, considering building everything in housing, raising capital to the network of chip foundry ies. But now IT appears that they've been in those plans do to cost and time constraints. Instead, according to reuters sources, they are opting to focus on in house chip designs.

The plane will now involve partnering with broad commer T S M C from manufacturing while adding A M D alongside in video as a supplier. The first day M D installation will come through microsoft azure who are using the companies M I three hundred x chips. Those chips are not as performing as in videos h one hundreds in training, but have outperformed in influence benchMarks.

Sources said that OpenAI had already been working with broad com for months to produce their first inference chip. Currently, G P, U supplies been bottled by strong command for training chips, but there's a chance that inference will become the scarce resource as more A I applications are deployed. OpenAI have reportedly assembled a twenty person chip design team LED by top former google engineers Thomas moran, Richard ho.

Source is also said that the firm has cured manufacturing capacity from T S M C and twenty six OpenAI C F O tera fier reference the growing pains that the company learn create its own data centers and chips during an interview earlier in the week stating it's definitely a stretch from a capital perspective, but also my own learning, Frankly, we are all learning in this space infrastructure is destiny. Sales force, which is making a huge bet on agents, has finally released them broadly. Beer, the new agent development platform, agent force to the public, the product is essentially a lower no code way of building and deploying agents.

C chap hots for employees and customers salesforce outline the key use case in a post highlighting the upgraded functionality. Your customer at midnight asks, I need to change my entire order. Traditional bot say, i've created a ticket, please wait 244 eight hours agent force service agent order， updated shipping rerouted confirmation sent anything else in a press release, the company said.

Agent force doesn't depend on human engagement to get work done. These agents can be trigger by changes in data. Business rules are prebuilt automation. Now sales for C E O mark ending off has recently become the leading voice, saying that the assistance error was all B S, and that everything is about agents, which is clearly marketing, given where they sit in a way to poke at microsoft.

But still, I think pretty exciting to start to see these agents actually rolling out in a real way will soon be able to see a lot more around how well they work in practice. That, however, is going to do for today's a ideally brief headlines edition. Next up, the main episode, today's episode is brought to you by Venus.

Venus is a private, uncensored r generate A I APP IT access is open source models to enable text, image and and co generation without the fear of being spied on or having your data exploded. Discuss anything with Venus without concern about IT being monitored, sold or given to advertisers in governments, Venus is different because your conversations and creations are kept curly within the browser, never stored or accessible by Venus, unlike other AI apps. Then I won't tell you what's OK to say or not.

Then is one patronized. You IT simply provides direct access to machine intelligence, no topics or off limits, no ideas or taboo. With Venus, your in control of the A.

I, as you should be, prosued ripes are available for forty nine dollars a year, or eight dollars per month. A ideally brief listeners receive a twenty percent discount on Venus pro, visit vena ice h nw and enter the discount code N L W daily brief. That's N L W daily brief, all one word.

Today's episode is brought you by super intelligent. Every single business workflow and function is being remade and reimagine with artificial intelligence. There is a huge chAllenge, however, of going from the potential of A I to actually capturing that value.

And that gap is what super intelligence is dedicated to. Filling super r intelligent accelerates AI adoption and engagement to help teams actually use A I to increase productivity and drive business value. An interactive AI use case registry gives your company full visibility into how people are using artificial intelligence right now.

Pare that with capabilities, building content in the form of tutorials, learning pats s and a use case library. And super intelligence helps people inside your company show how they're getting value out of A I, while providing resources for people to put that inspiration into action. The next three teams that sign up with one hundred or more seats are going to get free embedded.

That's a process by which our super intelligence team sits with your organization, figures out the specific use cases that matter most to you, and helps actually ensure support for adoption of those use cases to drive real value. Go to be super diet eyes. To learn more about this, A I enable ment network.

And now back to the show walk. Go back to the a daily brief. Today we are talking about what is an absolutely insane statistic. Yesterday, google reported correctly earnings, and on the analyst call, google CEO and arctic. I was asked effectively about the effectiveness of A I how much IT was real, how much IT was.

And one part of his response was that he said that more than a quarter of all new code at google is being generated by A I, then reviewed and accepted by engineers. Now there was a lot of other stuff on this call as well. A ee, for example, is helping drive revenue at google.

Google's cloud business, which includes its A I infrastructure products, was up thirty five percent year over year to eleven point four billion. But this stat around how much google is dog footing A I was really the headliner, but I went on to say that A I for coding was, quote, boosting productivity, efficiency and that this helps our engineers do more and move faster. I'm energized by our progress in the opportunities ahead, and we need to be laser or focused on building great products.

Of course, as business insider points out, as impressive as the number is, there may be some who get a little bit nervous hearing IT the new data from pitch. I will surely have some employees wondering whether they are coding themselves out of a job. Company leaders have previously promised that A I isn't taking google jobs yet, but the over twenty five percent figure is striking and underscores the benefits of improving this technology.

A lot of the discussion on x slash twitter was around the implication for developers audit chat rights. If you a programmer hesitant about using A, I, and trust me some, r, starting to look like resisting IT might mean getting left behind. Ron paul rights connecting the dots google laid off its entire python e in April four is part of a cost cutting effort today.

It's revealed on the earnings call that more than twenty five percent of all new code that google is generated by AI software engineering will be disrupted in more ways by AI than we can see right now. He also went on to make a prediction, I think that the next ten years is the last period that any human will ever write any code. Others try to connected to their own industries.

Lawyer Daniel lin, a junior, said, if more than twenty five percent of new code that google is generated by A, I and then reviews by engineers, I find IT quite plausible that the same as possible in law. If we do the work to build capable systems, there are differences in the ecosystems, and we need to close those gaps. There s nb r rights.

First A W S. And now google are automating large amounts of their code generation with the use of gena I software. Engineers that don't ever J I extensively have one, maybe two years of shelf life.

Investor Stevens in osi, I made the point that ultimately, users don't really care around how products were built. They just care that they work, he wrote in the early days of the microcomputer software was written in assembly, then see guide introduced to the micro. And all the industry was asking a, each vendor was, well, you moved to see in when that was because he was viewed as faster and easier to use and created less bugs.

IT was a modern, higher level language. So in the biggest vendors announce that their next versions would be in c. This happen again in one thousand nine and nine years old, with the rise of object oriented.

Soon inventors were claiming that the products were oo, that's some beautiful. All the magic beans came from that they will be easier to maintain, have your bugs easier to add new features. Note that wasn't true either than everyone to ended out those techniques.

No one cares what tools companies used to write code, not consumers or even enterprise. They care about features, cost, quality, security, reliability, privacy and performance, to name a few. If new tools helping, great. If new tools don't help them, that's not good.

It's even worse if the companies are out there are touting new tools and not delivering still raval to nucleus summed up the point of and dark comment saying this is to wall street, we are getting leverage on the ice spending. IT pretends what less sophisticated tech companies can accomplish over time. IT implies our eyes happening and gives a reason why they should keep spending on IT.

I will take a minute here to get up on my bully poppet once again and to talk about the two phases of general AI adoption. Once again, there is inevitably going to be a phase where companies treat AI as exclusively and efficiency technology, a way to get the same output with less input, a cost saving technology that will, in many cases, lead to job cuts, lower overall costs and the Marks are likely to rewarded, at least in the short term. However, the companies that win in the generate A I era, i'm quite convinced, will be those who view generate A I as an opportunity creation technology away.

In other words, to do more with the same or much, much more with just a little more when IT comes to developers, I am firmly on the other side of in ten years, we won't have any humans writing any code or at least them on the other side. When IT comes to humans producing code, I think that we are going to have hundreds and hundreds of thousands of times the code that we have now. I think more people are going to be producing code with the help of both assistance in agents.

And I think that means more things are going to be built. But in the in between, it's gonna get weird. And even as someone who watches this everyday, hearing that google is already generating a quarter of its cota eye is fairly surprising. Two other interesting google markets from the earnings call, one small and one bigger. The small one is that google says that they won't ship agents features until next year at the earliest.

The feature, known as project asha, was previewed at the I O developed conference in may and encompasses range of functionality, including a smart phone can recognize the world through the camera to A I assistance that can complete tasks autonomously on that same earnings call, eos and dark chai said google is building out experiences where A I can see in reason about the world around you, project astra is a glimpse of that future, were working to ship experiences like this is at least twenty twenty five. You might remember that the information had previously reported the google planning to ship their first agents, known as java, as early as december. But now that time, mine seems a little bit unclear.

Maybe the bigger deal is that microsoft get hub copilot will now support models from anthropic and google alongside OpenAI users will be given the choice of model between cloud three five senate gi one point 5 pro as well as GPT four O A one preview and a one mini github CEO Thomas dong key said there is no one model to rule every scenario， and developers expect the agency to build with the model that work best for them. IT is clear. The next phase of a eco generation will not only be defined by multimodal functionality, but by multi mode choice.

Microsoft, of course, introduced to have copilot 1 making IT one of the first products to demonstrate the power of A I assistance。 IT had previously lied solely on models provided by OpenAI, the rival of open the eyes of one models. Let get hub to explore the idea of a drop down menu to provide easy access to model options.

Donkey said at that point that felt the right time to expand other companies as well, he added. We're planning on extending that list in the future, but have no partner ships to announce at this point. Git has also introduced new automated code review feature, and next up on the feature list is a powerful APP designing to a cult Spark.

The feature will allow developers to create a prototypes based on text prompt, then refine the designs from there. Donkey said. For too long, there has been an unscalable barrier, eventually separating the vast majority of the worlds population from building software.

With Spark, we will enable over one billion personal computer and mobile phone users to build and share their own microbes directly on gets hub. Now of course, the subtext and context of this announcement is both one what appears to be a dentil, fraying relationship between microsoft and OpenAI. And too, of course, the battle between OpenAI and anthropic models when IT comes to coaster stance, develop nick topos rights, get up copilot adopted in law and gemini is the final nail in the coffin.

For base model companies, A I infrastructure is a difficult business to compete, even if they are producing amazing technology. Application layer wins rappers all day, everyday. Super fascinating.

I don't really know how much to read into this when IT comes to that open eye relationship peace IT certainly does put evidence in the column of those who think that based level models are going to become modified. Ed, matter what? It's a pretty big shift and one that i'll be worth watching over time. For now, though, that is going to do for today's ideally brief until next time, peace.