OpenAI

The Korea Herald reported today that OpenAI is in talks with Samsung to partner on a number of AI features that could be included in Galaxy phones. Citing people familiar with the matter, the report says that these discussions reflect a recent agreement between OpenAI and Apple, where OpenAI's technology powers Apple Intelligence features. The report mentions that AI giant OpenAI wants to position itself as a potential challenger to Google. Google has maintained its position as the leader in the browser and search market and has worked hard to strengthen its ...

In the early hours of Wednesday morning, OpenAI Chief Product Officer Kevin Weil confirmed via the X platform that Advanced Voice Mode will be fully available on the web version of ChatGPT for paid users this week. OpenAI first demonstrated ChatGPT's voice capabilities back in September. At that time, however, the feature was limited to iOS and Android apps. The advanced voice mode relies on the latest GPT-4o model with native audio...

OpenAI, in conjunction with nonprofit Common Sense Media, announced a free teacher training course designed to help teachers better understand AI and prompt engineering, the engineering of designing and optimizing prompts to guide a large language model to generate more accurate and useful output. The training course is geared toward teachers in grades K-12 and shows them how to use ChatGPT in a variety of educational use cases, such as creating course content or streamlining departmental meetings. The course is available on Common...

German music rights agency GEMA and India's ANI news agency filed lawsuits in their respective local courts on the 13th and 18th of this month, accusing OpenAI of using their content to train artificial intelligence models without permission. GEMA says it is the world's first class action organization to file a lawsuit against a provider of generative AI systems. The organization represents the copyrights of more than two million music rights holders worldwide, and is one of the largest associations of writers of musical works in the world. GEMA alleges that OpenAI's chat...

New evidence documents added to Musk's legal battle with OpenAI show that OpenAI, the star AI modeling company, considered getting into the AI chip business back around 2017, according to foreign media outlet TechCrunch. In a September 2017 email to Musk and OpenAI CEO Sam Altman, OpenAI co-founder and former chief scientist Ilya Sutskever said: "If we...

The original co-founder of OpenAI - IlyaSutskever noted that the training phase of using large amounts of unlabeled data to understand language patterns and structures is nearing its end. He mentioned that the results of extended training have plateaued, meaning that the method of boosting the performance of AI models by adding more data and arithmetic power (i.e. Scaling Law) has hit a bottleneck. Large Language Models (LLMs) like ChatGPT Current scaling strategies have reached their limits According to Ilya Sutskever, simply scaling by adding more data and computational resources...

OpenAI has announced the release of the desktop version of its ChatGPT app for all Microsoft Win10/11 users. In a press briefing, Alex Embiricos, product developer at OpenAI, said, "Users can easily start new conversations at any time without interrupting existing workflows, and can pick up unfinished conversations or ask new questions." The desktop version of ChatGPT supports both photo recognition and advanced voice, allowing users to communicate with their voice and quickly access ChatGPT with the Alt + Spacebar...

Nov. 14 (Bloomberg) -- U.S. artificial intelligence (AI) startup OpenAI recently called on the U.S. and its allies to work together to establish a "North American Artificial Intelligence Consortium" to support the infrastructure needed to develop AI systems to counter competition from China. OpenAI said Wednesday that the U.S. and its neighboring countries could form a "North American AI Alliance" to streamline access to talent, financing, and supply chain resources in order to promote the development of AI technology. The company noted that the model is expected to expand into a global network covering more U.S. allies and partners in the future, including Middle Eastern countries...

In a blog post yesterday (November 14), tech media outlet MacRumors reported that the desktop version of the ChatGPT app has introduced a new feature that works with programming tools such as Xcode, VS Code, TextEdit, and Terminal to streamline developers' workflow. ChatGPT provides access to Xcode and other apps through the Work with Apps feature, which allows developers to send code snippets to ChatGPT along with prompts...

The Information reported over the weekend that the rapid development of generative AI models appears to be hitting a bottleneck. Some experts are predicting that simply increasing model parameters, training data, and arithmetic power to improve model performance is becoming ineffective, and may even have negative consequences. OpenAI's latest flagship model, Orion, is a case in point. It has been reported that Orion's performance improvement over GPT-4 is not significant, and in some areas such as coding, there may not be any improvement at all. Safe Superintellig...

In a blog post today (Nov. 14), tech media outlet The Verge reported that OpenAI plans to launch an automated AI intelligence called "Operator" in January 2025, in the form of a research preview and development tool that can independently control computers and perform a variety of tasks. OpenAI's "Operator" isn't the only autonomous AI intelligence out there, as rival Anthropic recently unveiled its "computer use" feature, and Google is preparing to launch its "Operator" in ...

According to Bloomberg, AI giants including OpenAI, Google and Anthropic are facing "diminishing returns" as they hit a bottleneck in developing more advanced AI models. OpenAI's newest model, Orion, reportedly struggled with coding tasks, with no significant improvement over GPT-4. Google's upcoming Gemini software faces similar challenges, while Anthropic has delayed its highly anticipated Claude 3.5 o...

OpenAI's next-generation flagship model, Orion, may not have improved as much as its predecessor, according to The Information. Internal testers found that while Orion outperformed existing OpenAI models, the improvement was small compared to the leap from GPT-3 to GPT-4. This means that the pace of AI progress may be slowing. In fact, Orion may even be inferior to previous models in some areas, such as coding. To address this challenge, OpenAI set up...

November 11 news, local time on the 9th, OpenAI bet on the 1X company showed its humanoid robot Neo Beta's new skills -- cooking. Neo and Netflix chef Nick DiGiovanni had a home cooking showdown, competing to make a perfect medium rare steak.IT Home learned from public information that Nick is the youngest finalist in the history of MasterChef (MasterChef), and his YouTube channel currently has 21.2 million followers! ...

Nov. 9 (Bloomberg) -- Lilian Weng, OpenAI's other chief security researcher, announced Friday that she is leaving the startup, according to a report from FT. Weng has served as vice president of research and security since August, and prior to that she was head of OpenAI's security systems team. In a post on X, Ms. Weng said, "After seven years at OpenAI, I feel ready to reboot and explore something new." Ms. Weng said her last day will be November 15, but did not specify...

OpenAI launched a direct challenge to Google last week with the official release of its highly anticipated search product, ChatGPT Search. The timing of this announcement was long anticipated, with Google incorporating AI-generated answers into its core product earlier this year, but there were some embarrassing "hallucinations" along the way. This mistake led many to believe that OpenAI's search engine would be a real "Google killer. However, after using ChatGPT Search as the default search engine for a day or so...

OpenAI CEO Sam Altman has always been passionate about general-purpose artificial intelligence (AGI) and says the company is actively pursuing this ambitious goal. In a recent Reddit AMA, Altman claimed that AGI is possible using existing hardware, and while he didn't specify exactly what would be needed, he has previously said that realizing his AI vision would require $7 trillion and years to build 36 semiconductor factories and more data centers. Altman previously wrote in a blog post...

OpenAI CEO Sam Altman poured cold water on hopes for the next major release of ChatGPT, due out this year. "We're going to release some really good versions later this year! We won't be calling it gpt-5, though," he said at this week's Reddit AMA. OpenAI will focus on releasing GPT-o1, previously codenamed "Project Strawberry," a new AI model from OpenAI that is slower but aims to answer before...

OpenAI announced early this morning that it has added a new search function to its big model product ChatGPT, officially entering the search space long dominated by Alphabet's Google, as well as competing with Microsoft's Bing and Perplexity, an emerging search-based AI chatbot service. OpenAI says ChatGPT's search function automatically decides whether or not to search the web based on a user's needs, and can quickly provide up-to-date information with links to relevant web pages. The official blog says ChatGPT...

On October 31, OpenAI announced that it is open-sourcing a new benchmark called SimpleQA, which measures the ability of language models to answer short fact-seeking questions, in order to measure the accuracy of language models. One of the open challenges in AI is how to train models to generate factually correct answers. Current language models sometimes produce incorrect output or unsubstantiated answers, a problem known as "hallucinations". Language models that can generate more accurate and less illusory answers are more reliable and can be used...

October 31, 2011 - OpenAI today announced that ChatGPT's Advanced Voice Mode (AVM) is now available on Windows and Mac platforms, running on OpenAI's state-of-the-art GPT-4o model. Based on OpenAI's state-of-the-art GPT-4o model, AVM allows users to communicate with ChatGPT as if they were having a conversation with a real person, including interruptions, pauses, and other natural language behaviors.OpenAI says AVM provides a more natural, real-time conversational experience, with the ability to sense and respond in real time to...

According to Bloomberg, about 75% of OpenAI's revenue comes from consumer subscriptions, a figure revealed by CFO Sarah Friar. OpenAI's consumer strategy is centered on the ChatGPT product, with monthly subscriptions starting at $20 (currently around Rs. 143). According to the report, OpenAI has successfully converted free users from 5% to 6% into paid members. Currently, ChatGPT has 250 million active weekly users. In addition to the consumer market, OpenAI is also highly...

More than a dozen software engineers, developers, and academic researchers say OpenAI's voice transcription tool Whisper has a major flaw: it sometimes generates large paragraphs or even entire sentences of false content out of thin air, according to a report by the Associated Press on 27 May. The experts note that the generated text, commonly referred to in the industry as "AI hallucinations," can involve racial remarks, violent language, and even fabricated medical advice. This is particularly worrisome, according to the experts, because Whisper is already being used in a wide range of industries around the world, including for translating and transcribing interviews...

October 25, 2012 - Yesterday afternoon, OpenAI CEO Sam Altman responded to recent reports of a "next-generation model called Orion" on the X platform: fake news out of control. Previously on X-Platform, Sam Altman, CEO of OpenAI, responded to recent reports of a "next-generation model, Orion," saying "fake news out of control. "Orion" adopts a different release model from GPT-4o and o1, and will not be released via ChatGPT ...