Google is reportedly working on Apprentice Bard, a ChatGPT competitor
Apprentice Bard is said to utilize the company's LaMDA technology to compete with ChatGPT's back-and-forth discussion format.
The race for true artificial intelligence continues with Google being the latest company attempting to solve the various problems the technology presents. A recent report from CBNC shows that Google has its eyes on ChatGPT. As such, Google is reportedly using its LaMDA technology in an AI program known internally as Apprentice Bard.
In documents and designs viewed by CNBC, it appears that Google is currently experimenting with various AI programs in an attempt to compete with ChatGPT. The OpenAI chat program has garnered a lot of attention recently, with Microsoft backing the team to the tune of several billion dollars. Google’s attempt at competition includes its own chat program and experiments with its web search function.
The report by CNBC goes into some detail about a program called Apprentice Bard. This program uses Alphabet’s Language Model for Dialogue Applications (LaMDA) to allow users to ask questions, receive information, and provide feedback. This is similar to how ChatGPT works. One standout difference is that Apprentice Bard can pull information from recent events.
In one example, an employee that was testing the program asked whether Google would perform more layoffs. The AI program says it is unlikely, especially considering “Google’s revenue increased by 34% in 2021, and the company’s stock price has risen by 70% since January 2022.” Despite these impressive figures, Google still laid off 12,000 employees at the end of January.
Google’s interest in creating an AI competitor makes sense. With companies actively pursuing and researching artificial intelligence, it’s an emerging field that provides untold potential for various industries. Recently, ChatGPT successfully passed an MBA exam while back in 2019 OpenAI beat Dota 2 champs in a best-of-three match.
Part of AI advantages includes helping users get information in a new format. To that end, Google is reportedly experimenting with AI in its web search functionality. CNBC acquired documents that showed Google’s search bar featuring various question prompts, chat logos inside the search bar, and more conversational answers to queries.
With Microsoft, NVIDIA, Tesla, and many more companies entering the realm of artificial intelligence, it only makes sense for Google to throw its hat into the ring. Be sure to keep it locked to Shacknews as we bring you the latest information about AI.
-
Sam Chandler posted a new article, Google is reportedly working on Apprentice Bard, a ChatGPT competitor
-
Google is asking employees to test potential ChatGPT competitors, including a chatbot called ‘Apprentice Bard’
https://www.cnbc.com/2023/01/31/google-testing-chatgpt-like-chatbot-apprentice-bard-with-employees.html
*Google is testing ChatGPT-like products that use its LaMDA technology, according to sources and internal documents acquired by CNBC.
* The company is also testing new search page designs that integrate the chat technology.
* More employees have been asked to help test the efforts internally in recent weeks.
How long before Google essentially just recreates 'Ask Jeeves'? Anybody remember 'Ask Jeeves'?-
-
-
It’s very hard for a company to disrupt itself. It’s a classic business school thing. You make money doing X, and someone in the company says “we could make more money doing Y!” And the internal response is “if we put money toward Y, that takes away money we could put toward making X better.” The net result is that Y has to come from another company.
-
OpenAI is certainly more nimble than Google. But when the public starts talking about the entire core business model of a tech giant being in serious danger, very powerful people within Alphabet are going to take notice. OpenAI will probably continue to be more experimental than Google, but search is Google's game to lose and they know it.
-
An example that comes to mind, but in reverse, is Google Fiber. They were a scrappy startup that created some real disruption in the industry. They spooked the big players. But then the big players just went and did what they were doing, but at a vast scale, completely overshadowing the whole reason for Fiber to exist.
-
-
What can OpenAI do that is almost impossible for Google to replicate? They made a smart play with ChatGPT and gained a lot of name recognition as a result, sure. But Google is completely focused on this now. They've been focused on AI for years, they have all the infrastructure in place already, they just didn't release this specific product first. If they can put out a competitor this year, can you really say they were disrupted?
-
When Google releases Android as a mobile operating that is free for OEMs to use, or even paying OEMs to use it (via a share of on device search revenue) that is disrupting Windows Mobile. The problem is not that Microsoft literally cannot make a mobile OS and give it away. It’s that the entire business is oriented in a way that a) that strategy would never occur to or be palatable to the company previously and b) is still impossible to compete with (there is no Bing Ads search revenue to share that can compete with Google Search Ad share).
Disruption is about business models, not product features.
-
-
-
-
-
Imagine an evolution of ChatGPT that is more accurate, shows its sources, and has up to date information from the internet so that you would actually start using it instead of Google for all sorts of general purpose searches. Could Google technically make such a product or something close to it? Presumably. But how will it make money? Where are they going to stick ads? What search terms are you going to sell ads against? No one wants a ChatGPT that first suggests an ad or interleaves ads in its responses (nevermind how you’d even technically implement that). It’s a threat to the entire way Google makes money. Google is not a highly diversified company. Google Search, search ads, and the related products/value of that (ie data they learn to better target other ads) drive the vast majority of the company. They can’t just launch a product that makes the revenue of Google Search go to $0. Nor can they attempt to suddenly turn every Google Search user into a subscriber. This is a similar threat that voice assistants like Alexa represented but the AI was not advanced enough to actually be a general purpose search via natural language.
The other major difference is reputational. Little companies can take risks big companies cannot. If ChatGPT accidentally writes some pro holocaust content when prompted it generates some news in the tech press and people move on. No Congress members or regulators will care. Whereas if Google Search starts serving pro holocaust content at the top of search results it might make headline news and draw the ire of legislators.
So while ChatGPT might light a fire under some parts of the company the fundamental pillars that lead them not to launch a similar product in the first place still exist. Might they take slightly more risks now (especially reputationally)? Yes. But the business model problem is the more fundamental issue.-
-
I’m sure you could prompt it to do this with ease. But that’s not the problem. The problem is whether a competitor will do this. If Google has a version that interleaves ads, and a competitor doesn’t (because their existing business isn’t beholden to a legacy ad business) then which product will you use?
-
-
-
-
Man Ask Jeeves was like the shittiest search engine but the marketing was incredible. People just liked it.
The sheer number of times I’ve seen someone talk about their site’s search capability as “this isn’t Ask Jeeves, so…” as if Ask Jeeves was the benchmark.
Sort of like how Geek Squad is a horrible service but people love the name and the marketing and the little VW bugs they at least used to drive and I hear people all the time say “guess I should call Geek Squad?”-
https://productmint.com/what-happened-to-ask-jeeves/#:~:text=Ask%20Jeeves%2C%20now%20called%20Ask,position%20in%20the%20search%20industry.
Warthen, prior to launching the search site, was the co-founder of a software development company that did projects for the likes of Microsoft or Logicraft, amongst others.
He eventually became acquainted with Gruener who, at the time, was working as a venture capitalist and, therefore, had a front-row seat into how internet businesses were taking off.
At the time, search was dominated by AltaVista, which was launched in December 1995. Although AltaVista represented a vast improvement compared to the search engines that were out there, its search functionality was dominated by simple one to two-word combinations.
Gruener and Warthen figured that internet users would want to search the web by being able to use natural language. Results, as previously stated, would then be supported by tropes of human editors.
In June 1996, the two officially incorporated the company. To highlight the search engine’s personal touch, they took inspiration from the work of P. G. Wodehouse. His fiction entailed a valet named Jeeves who served the English gentleman Bertie Wooster.
After close to a year of working on the business, Ask Jeeves was unveiled to the public on June 1st, 1997.-
I meant to add that it eventually failed.
Ask Jeeves was shut down because Ask.com, the company behind it, wanted to undergo a rebrand to further enhance its position in the search industry.
At the time of the shutdown, right after Ask Jeeves was acquired by IAC, the company had made it known that it wanted to become a stronger competitor to Google.
As a result, it invested heavily into advertising, skimmed down the ads on its site, and made some hefty acquisitions.
In order to be associated as a search and not a Q&A brand, the executives figured that it was best to get rid of the butler.
After all, Ask Jeeves did manage to acquire a lot of customer goodwill and, as a result, had a strong brand associated with it.
Certainly, the company did conduct tons of interviews and research to ensure that the brand switch was the correct strategic move.
In the end, it wasn’t enough to compete with the likes of Google, Bing, and others who all simply were too far ahead to be caught.
-
-
-
-
-
-
-
-
-
-
-
The point is competitors don’t have the same requirement. If a competitor finds a way to monetize without ads then which product will you use? The one that gives you instant answers or the one that makes you watch a pre roll video? Maybe ads do end up the way to monetize it but only Google is boxed in to a) needing it to be ads and b) needing those ads to monetize as well as Google Search ads.
-
-
-
-
-
-
-
-
ChatGPT sets record for fastest-growing user base in history, report says
"On Wednesday, Reuters reported that AI bot ChatGPT reached an estimated 100 million active monthly users last month, a mere two months from launch, making it the "fastest-growing consumer application in history," according to a UBS investment bank research note. In comparison, TikTok took nine months to reach 100 million monthly users, and Instagram about 2.5 years, according to UBS researcher Lloyd Walmsley.
Further Reading
Fearing ChatGPT, Google enlists founders Brin and Page in AI fight
“In 20 years following the Internet space, we cannot recall a faster ramp in a consumer internet app," Reuters quotes Walmsley as writing in the UBS note.
Reuters says the UBS data comes from analytics firm Similar Web, which states that around 13 million unique visitors used ChatGPT every day in January, doubling the number of users in December.
ChatGPT is a conversational large language model (LLM) that can discuss almost any topic at an almost human level. It reads context and answers questions easily, though sometimes not accurately (improving its accuracy is a work in progress). After launching as a free public beta on November 30, the GPT-3 powered AI bot has inspired awe, wonder, and fear in education, computer security, and finance. It's shaken up the tech industry, prompting a $10 billion investment from Microsoft and causing Google to see its life flash before its eyes.
Also on Wednesday, OpenAI announced ChatGPT Plus, a $20 per month subscription service that will offer users faster response times, preferential access to ChatGPT during peak times, and priority access to new features. It's an attempt to keep up with the intense demand for ChatGPT that has often seen the site deny users due to overwhelming activity.
Over the past few decades, researchers have noticed that technology adoption rates are quickening, with inventions such as the telephone, television, and the Internet taking shorter periods of time to reach massive numbers of users. Will generative AI tools be next on that list? With the kind of trajectory shown by ChatGPT, it's entirely possible."
Source: https://arstechnica.com/information-technology/2023/02/chatgpt-sets-record-for-fastest-growing-user-base-in-history-report-says/ -
-
-
-
-
-
Speaking of Siri... I have no idea really what it can do. Something I really miss on my Windows Phone was the ability to get reminders when I was at certain locations.
So like, I could tell my WP "remind me to get toothpaste when I'm at a pharmacy" and it would remind me whenever I was walking by a pharmacy.
Can Siri do something like that?
-
-
-
ChatGPT Plus, $20/mo: https://openai.com/blog/chatgpt-plus/
I think a big question with this tech is whether it will ever be affordable to offer it for free. ChatGPT right now is doing the loss leader thing. They'll probably have to clamp down on the sophistication of the AI they offer to free users in comparison to paid users. -
-
-
-
-
a slightly off topic google idea is that in retrospect, the Alphabet "unspinoff" was a total mistake.
Alphabet with it's "moonshots" should have been the ones to buy out OpenAI. Back then, the theory was that Alphabet would be the holding space for not-as-profitable-but-still-needed business units.
All of the other Alphabet business units that are not Google related have died/sunsetted/had 0 impact.
Alphabet itself is just overhead. If they fired every single alphabet executive, and just used google executives, it would be the same.
-