The impact of AI on OpenAI: OpenAI’s opentext service and its collaboration with DeepMind and Google in the 21st century
ChatGPT stands out because it can take a naturally phrased question and answer it using a new variant of GPT-3, called GPT-3.5. The new capacity to respond to all kinds of questions gives the model a compelling new interface just about anyone can use. The openai service was open for free and the fact that it can have good fun also helped fuel the chatbot’s viral debut, similar to how some tools for creating images using artificial intelligence have proven great for meme making.
More than that, artificial intelligence systems have for years underpinned many of the functions people may now take for granted, from content recommendations on social media platforms and auto-complete tools in e-mail to voice assistants and facial recognition tools. The power of artificial intelligence was on full display for millions in an entertaining and immediately graspable way when it was publicly released in November. ChatGPT simultaneously made it much easier to see how far the technology has progressed in recent years and to imagine the vast potential for the impact it could have across industries.
OpenAI has not released full details on how it gave its text generation software a naturalistic new interface, but the company shared some information in a blog post. The team fed human written answers to GPT- 3.5 as training data, and then used a form of reinforcement learning to make the model give better answers to the questions.
“We are absolutely looking to get these things out into real products and into things that are more prominently featuring the language model rather than under the covers, which is where we’ve been using them to date,” said Dean. It is important that we get this right. Pichai added that Google has a “a lot” planned for AI language features in 2023, and that “this is an area where we need to be bold and responsible so we have to balance that.”
One of the reasons why the situation may be particularly vexing to some of the company’s experts is that researchers developed some of the technology behind the project. “We re-oriented the company around AI six years ago,” Pichai wrote. We have continued to invest in AI across the board. He name-checked both Google’s AI research division and work at DeepMind, the UK-based AI startup that Google acquired in 2014.
OpenAI used to be cautious in its development of its LLM technology, but changed its stance with the launching of the CHATGPT, offering wide access to the public. The result has been a storm of beneficial publicity and hype for OpenAI, even as the company eats huge costs keeping the system free-to-use.
In early February, first Google, then Microsoft, announced major overhauls to their search engines. Both tech giants have spent a lot of money on building or buying generative technology. They are hoping to give users a richer, more accurate experience by integrating them into search. The Chinese search company has said it will do the same.
For example, the query “Is it easier to learn the piano or the guitar?” would be met with “Some say the piano is easier to learn, as the finger and hand movements are more natural … Others say it is easier to play the guitar. Pichai also said that Google plans to make the underlying technology available to developers through an API, as OpenAI is doing with ChatGPT, but did not offer a timeline.
There is a lot of enthusiasm for the launch of a new search engine, and we believe that it will pose a serious challenge to its dominance of the web search market. Microsoft, which recently invested around $10 billion in OpenAI, is holding a media event tomorrow related to its work with ChatGPT’s creator that is believed to relate to new features for the company’s second-place search engine, Bing. Sam Altman took a picture of himself with Microsoft’s CEO, a few minutes after the announcement.
Adding technology behind LaMDA to products is something that has been decided to be cautious about, by its own admission. Artificial intelligence models that are trained on text from the Web are prone to displaying racist and sexist biases.
Those limitations were highlighted by researchers in a draft research paper in 2020, which led to the firing of ethical artificial intelligence researchers, Timnit Gebru and Margaret Mitchell.
Some of the researchers working on the LaMDA project left the company because of their hesitancy about using the same technology that they worked on. The company may have been inspired to speed up the time it takes for it to add text generation capabilities to its products.
Google is expected to announce artificial intelligence integrations for the company’s search engine on February 8 at 8:30 am Eastern. You don’t need a subscription to watch live on YouTube.
Is generative intelligence ready to help you surf the web? These models are costly to power and hard to keep updated, and they love to make shit up. Public engagement with the technology is rapidly shifting as more people test out the tools, but generative AI’s positive impact on the consumer search experience is still largely unproven.
You may be familiar with AI text and AI images, but these mediums are only the starting point for generative AI. More information about the research into possibilities for audio and video can be found on the internet. As mainstream uses for large language models emerge, a lot of startups in Silicon Valley are also vying for attention.
Microsoft executives said that a limited version of the AI-enhanced Bing would roll out today, though some early testers will have access to a more powerful version in order to gather feedback. A wider-ranging launch will happen in the coming weeks, so the company is asking people to sign up.
The new Bing interface is powered by openai technology that learns to generate text by analyzing the statistical pattern of words in articles, forums, and other content on the internet.
The response also included a disclaimer: “However, this is not a definitive answer and you should always measure the actual items before attempting to transport them.” A “feedback box” at the top of each response will allow users to respond with a thumbs-up or a thumbs-down, helping Microsoft train its algorithms. A demonstration of the use of text generation to enhance search results occurred yesterday.
The factual error highlights the importance of a rigorous testing process that is being implemented this week by the Trusted Tester program. Bard will use external feedback with their own testing to make sure their responses meet a high bar.
A user asked Bard if he could tell his 9 year old about new discoveries from the james whackle space telescope in the demo. Bard responds with a series of bullet points, including one that reads: “JWST took the very first pictures of a planet outside of our own solar system.”
According to NASA, however, the first image showing an exoplanet – or any planet beyond our solar system – was actually taken by the European Southern Observatory’s Very Large Telescope nearly two decades ago, in 2004.
Artificial Intelligence in Search Engines: a View from Alphabet, Google, Bing, Baidu and Google+Baydu
Shares for Google-parent Alphabet fell as much as 8% in midday trading Wednesday after the inaccurate response from Bard was first reported by Reuters.
In the presentation Wednesday, an executive teased the plans to use this technology to offer more complex and sophisticated responses to queries, as well as providing bullet points on the best times of the year to see different constellations and offering pros and cons for buying an electric vehicle.
Do you need to write a real estate listing or employee review? Your first draft is done in three seconds, if you insert a few words into the query bar. Want to come up with a plan for quick meal and groceries that are according to your sensitivities? Bing, apparently, has you covered.
Last but by no means least in the new AI search wars is Baidu, China’s biggest search company. It joined the competition with the announcement of a new rival, known as “Ernie Bot” in English. Baidu says it will release the bot after completing internal testing this March.
Alan Woodward, professor of cybersecurity at the University of Surrey in the UK, says that there are already a lot of resources involved in searching for internet content, but that it is not enough to incorporate Artificial Intelligence. “It requires processing power as well as storage and efficient search. The power and cooling resources of large processing centres are increased when a step change in online processing occurs. I think this could be a big step.
Microsoft’s Bing search engine will be powered by large language models, such as those that underpin Open Artificial Intelligence’s chatGPIt, which means parsed and computing links within massive volumes of data.
Carlos Gmez-Rodrguez is a computer scientist at the University of Corua in Spain. “Right now, only the Big Tech companies can train them.”
Third-party analysis estimates that the training of GPT3 led to emissions of more than 500 million tons of carbon, while neither OpenAI nor GOOGLE have said what their computing costs are.
“It’s not that bad, but then you have to take into account Gmez-Rodrguez says that it is necessary to execute it and serve millions of users.
There isn’t a lot of difference between integrating the ChatGPT product into Bing and using it as a separate product.
To meet the requirements of users of the search engine, that will have to change. It is a different scale of things if they add more parameters and retrain the model often.
What will we learn from the 10th anniversary of the birth of the Internet? The impact of AI on tech innovation in Silicon Valley and how it affects business
Executives in business casual wear trot up on stage and pretend a few changes to the camera and processor make this year’s phone profoundly different than last year’s phone or adding a touchscreen onto yet another product is bleeding edge.
This week, that changed in a different way. A number of some of the world’s largest companies teased big changes to their services, which are central to our everyday lives and how we experience the internet. In each case, the changes were powered by new AI technology that allows for more conversational and complex responses.
If the introduction of phones defined the 2000s, the 2010s in Silicon Valley was defined by the products that didn’t arrive, such as self- driving cars and virtual reality products that were cheaper but still not ready for use.
New generations of technologies are often not visible to the general public because they have yet to mature enough to be useful. When it is directly accessible to the public, and if you look over time, you can see them in an industrial setting or behind the scenes, then there is more public interest.
Some people worry it could disrupt industries, potentially putting artists, tutors, coders, writers and journalists out of work. Others are more optimistic, postulating it will allow employees to tackle to-do lists with greater efficiency or focus on higher-level tasks. It will likely force industries to change but that isn’t the case. It’s a bad thing.
“New technologies always come with new risks and we as a society will have to address them, such as implementing acceptable use policies and educating the general public about how to use them properly. Guidelines will be needed.
The tech companies are not paying enough to feed the public: What are the best dog beds? A question for the Microsoft Bing chatbots of Wirecutter
Two years ago, Brad Smith told a US congress that tech companies were not paying enough to media companies for the news content they used to give to search engines.
It is far bigger than us and we hope people do not use gadgets like cellphones or laptops in a century. Our democracy is dependent on it. Smith said tech companies should do more and that Microsoft was committed to continuing “healthy revenue-sharing” with news publishers, including licensing articles for Microsoft news apps.
When WIRED asked the Bing chatbot about the best dog beds according to The New York Times product review site Wirecutter, which is behind a metered paywall, it quickly reeled off the publication’s top three picks, with brief descriptions for each. “This bed is cozy, durable, easy to wash, and comes in various sizes and colors,” it said of one.
When the bot responded, it gave citations to Wirecutter’s reviews and websites that used his name to make money by using affiliate links. The Times was unresponsive to a request for comment.
Source: https://www.wired.com/story/news-publishers-are-wary-of-the-microsoft-bing-chatbots-media-diet/
The Bing API is not an Alternative to Search Engines: Why the Microsoft Adversarial Services (Bing) Describe the Interaction with Content Publishers
Microsoft communications director Caitlin Roulston says that “Bing only crawls content publishers make available to us.” She says that Microsoft has an agreement with some publishers that give the search engine access to paywalled content. The scheme predates Bing’s AI upgrade this week.
OpenAI is not known to have paid to license all that content, though it has licensed images from the stock image library Shutterstock to provide training data for its work on generating images. Microsoft does not pay content creators when its bot summarizes articles, just as it has not paid web publishers to display short snippets from their websites in search results. Bing’s interface provides richer answers than traditional search engines do.