ຂ່າວ

According to Decrypt, Apple has launched its latest software update, introducing a range of artificial intelligence features across its mobile and desktop operating systems. The update, which includes iOS 18.2, iPadOS 18.2, and MacOS Sequoia 15.2, aims to enhance user experience with new capabilities such as ChatGPT support for Siri, improved camera controls, personalized emoji design, and generative image creation. These features are available on iPhone 15 Pro and later, iPad with A17 Pro or M1 and later, and Mac with M1 and later. One of the standout features is the integration of OpenAI's ChatGPT into Siri, allowing users to access the chatbot's capabilities without needing an account. This integration enables users to generate text-based responses and images directly through Siri. Apple emphasizes its commitment to privacy, ensuring that requests sent to ChatGPT are not stored or used to train OpenAI's models. Additionally, the update introduces the Image Playground, a generative AI tool that allows users to create images based on their photo library, with themes ranging from animation to illustrations. This feature is accessible through various apps, including Freeform and Keynote. The update also introduces "genmoji," customizable emojis that can be generated via text descriptions or photos. These genmoji can be personalized with accessories and shared as stickers or reactions. Furthermore, the iPhone 16's Camera Control receives an upgrade with Visual Intelligence, offering instant object and text recognition. This feature allows users to copy or translate text, detect contacts, and search for items online, even using ChatGPT for explanations. Apple has also enhanced its TV and Music services, enabling users to find content by describing moods, genres, actors, or decades, making discovery more intuitive. In terms of privacy and security, Apple has taken measures to obscure IP addresses for users utilizing the new ChatGPT integration, preventing session linkage. For those connecting their ChatGPT Plus/Pro accounts, OpenAI's data-use policies will apply. Apple ensures that user data on its Private Cloud Compute is used solely to fulfill requests and is never stored or shared. Additionally, Safari's search capabilities have been upgraded, with URLs automatically converted to HTTPS to secure online activity.

According to Odaily, the global generative artificial intelligence (AI) market is anticipated to experience significant growth over the next five years. A recent forecast by Canalys indicates that the market, currently in its nascent stage, is expected to expand fivefold. The market value is projected to rise from $14.6 billion in 2024 to $72.8 billion by 2029.This substantial growth underscores the increasing adoption and integration of generative AI technologies across various industries. As businesses and organizations continue to explore the potential of AI-driven solutions, the demand for generative AI is likely to surge. 

According to Odaily, OpenAI has launched its fourth day of technical live streaming, unveiling a new feature called Canvas. This feature is integrated directly into ChatGPT, offering a separate window primarily designed for programming and writing tasks. Canvas provides suggestions, reviews, and executes specific functions, delivering more detailed services.Canvas acts as an embedded agent to assist and guide ChatGPT in generating and refining content more effectively. 

According to Odaily, OpenAI's Chief Financial Officer has stated that enterprise clients are expected to pay thousands of dollars for the use of artificial intelligence tools. This pricing structure is anticipated to be justified in the long run. The CFO emphasized that the costs associated with these AI tools reflect their value and the significant benefits they offer to businesses. As AI technology continues to evolve, companies are increasingly integrating these tools into their operations to enhance efficiency and innovation.The CFO's remarks underscore the growing importance of AI in the corporate sector, where businesses are willing to invest in advanced technologies to maintain a competitive edge. The financial commitment required for these AI tools is seen as an investment in future capabilities, enabling companies to leverage data-driven insights and automation. As AI adoption becomes more widespread, the pricing model is expected to align with the value delivered, ensuring that enterprises receive a return on their investment.

According to CoinDesk, the past year marked a significant transformation for bitcoin miners as many redirected their efforts towards establishing artificial intelligence (AI) data centers. Core Scientific, under the leadership of CEO Adam Sullivan, stood out in this shift. Emerging from bankruptcy in January, the company secured a multi-billion dollar deal with AI hyperscaler CoreWeave, a long-time business partner.Core Scientific was the pioneer among miners to venture into AI computation on a scale that reignited investor interest in the mining sector. The successful deal led to a substantial increase in the company's share price, prompting other miners to follow suit and integrate AI into their operations. This shift has now become a standard practice within the industry, with investors urging miners who have not yet diversified into AI to consider doing so.Given the strategic deal and the prevailing investor sentiment, it is anticipated that private equity firms or major tech companies, which are in constant need of data centers, might soon express interest in acquiring Core Scientific. This trend could potentially extend to other bitcoin miners as well, as the industry continues to evolve and adapt to new technological demands.

According to PANews, OpenAI has announced the official release of its AI video generation model, Sora, to users. This development comes ten months after the initial preview of the product. Sora is designed to create realistic videos based on text prompts and will be available to ChatGPT's paid users in the United States and other markets later in the day. The new tool, Sora Turbo, can generate videos up to 20 seconds long and offers various versions of these videos. OpenAI's CEO, Sam Altman, noted that the initial use of Sora will be limited. ChatGPT Plus subscribers will be able to generate 50 low-resolution videos per month, while ChatGPT Pro subscribers will have the capability to generate an unlimited number of high-resolution videos at a slower pace, along with 500 videos that can be generated quickly. This strategic release aims to enhance user experience by providing advanced video generation capabilities, marking a significant step in AI-driven content creation.

According to PANews, the Financial Times has reported that Meta is launching internal artificial intelligence tools aimed at enhancing its presence in the commercial sector. This strategic move is part of Meta's broader efforts to leverage AI technology to gain a competitive edge in the business market.

According to Odaily, market sources have revealed that OpenAI is in discussions with Microsoft to potentially remove the AGI (Artificial General Intelligence) clause from their agreements. This move is aimed at unlocking further investment opportunities. The discussions between the two tech giants indicate a strategic shift that could have significant implications for the future of AI development.The AGI clause has been a critical component of OpenAI's agreements, reflecting the company's cautious approach towards the development of advanced AI technologies. By considering the removal of this clause, OpenAI and Microsoft may be signaling a readiness to accelerate their AI initiatives and attract more investment. This development comes amid growing interest and competition in the AI sector, where companies are racing to advance their technologies and secure a leading position in the market.The potential removal of the AGI clause could pave the way for increased collaboration and innovation between OpenAI and Microsoft, as well as with other potential investors. As the discussions progress, the tech industry will be closely watching for any official announcements or changes in strategy from either company. This move could also influence other players in the AI field to reconsider their own approaches to AGI and investment strategies.

According to PANews, China's National Intellectual Property Administration has released a draft of the 'Guidelines for Invention Patent Applications Related to Artificial Intelligence' for public consultation. This initiative aims to assist patent applicants in better understanding the policies surrounding AI patent examinations, thereby improving the quality of applications and addressing key legal issues of interest to innovators.The draft guidelines are designed to provide clarity and guidance to those seeking patents in the rapidly evolving field of artificial intelligence. By outlining the examination policies, the guidelines seek to enhance the overall quality of patent applications and ensure that they meet the necessary legal standards. This move is part of a broader effort to support innovation and address the legal concerns that have emerged as AI technology continues to advance.The public is invited to submit their opinions on the draft guidelines until December 13, 2024. Feedback can be provided through various channels, including email, fax, or postal mail. This open consultation process reflects the administration's commitment to engaging with the public and stakeholders in shaping the future of AI-related patent policies in China.

According to PANews, prominent AI institutions Grass, Ontocord, and LAION have jointly announced the release of the VALID (Video-Audio Large Interleaved Dataset) dataset.This dataset is constructed based on Grass's video repository and includes 30 million audio segments. These audio segments are interleaved with images and text, making it the first video-audio interleaved dataset in the industry. The release of VALID is expected to provide new data support for the training of multimodal AI models.

According to Decrypt, Google CEO Sundar Pichai addressed the evolving landscape of artificial intelligence at the New York Times’ annual Dealbook Summit. Pichai suggested that the era of AI developers leveraging large datasets from the internet might be nearing its end. He noted that while a few companies have excelled with the current generation of large language models (LLMs), the path forward will likely become more challenging. This sentiment arises amid reports of a slowdown in AI model performance compared to the rapid advancements seen two years ago with the launch of ChatGPT.Pichai's observations align with concerns from other industry leaders, including Ethereum co-founder Vitalik Buterin and a16z's Marc Andreessen and Ben Horowitz. They have pointed out that scaling AI models with vast amounts of unlabeled data has reached a plateau. Pichai emphasized that by 2025, the "low-hanging fruit" will be exhausted, and only elite teams will distinguish themselves in the field. Despite these challenges, Pichai remains optimistic about future advancements, suggesting that significant progress is expected in the coming years.A notable concern in the AI community is the potential for an "AI ouroboros" effect, where models train on data generated by other AIs rather than human-created content. This phenomenon, named after the ancient symbol of a serpent eating its tail, could lead to repetitive or distorted outputs as systems increasingly rely on AI-generated data. While some developers worry about AI models reaching a plateau, Pichai believes that deeper breakthroughs will be necessary to advance to the next stage of AI development. He anticipates that 2025 will be an exciting year for the field, with significant progress expected despite the challenges ahead.

According to Odaily, OpenAI's founder Sam Altman announced the release of the full-featured inference model o1 and ChatGPT Pro, which comes with a monthly subscription fee of $200. The o1 model is described as the most intelligent model in the world, surpassing the capabilities of the previous o1-preview version. It offers enhanced speed, intelligence, and additional functionalities, such as multimodal capabilities. The o1 model is now available on ChatGPT and is set to be introduced to the API soon. The newly launched ChatGPT Pro subscription will allow users to fully leverage the model and its tools, including unlimited access to OpenAI's o1 and a Pro-exclusive version of o1. This development marks a significant advancement in AI technology, providing users with more powerful and versatile tools for various applications.

According to PANews, U.S. defense technology company Anduril has announced a strategic partnership with OpenAI to develop and deploy advanced artificial intelligence technologies aimed at supporting national security missions. This collaboration will integrate OpenAI's cutting-edge AI models with Anduril's high-performance defense systems and Lattice software platform to enhance the real-time threat detection and response capabilities of counter-unmanned aircraft systems (CUAS). The partnership aims to bolster the defense capabilities of the U.S. military and its allies against aerial threats, optimize data processing, reduce the burden on operators, and improve situational awareness. Both parties emphasize that this initiative is not only a response to technological competition but also a commitment to ensuring the safe and responsible use of AI technologies in the national security sector, while upholding the values of freedom and democracy.

According to PANews, OpenAI recently experienced a brief performance decline affecting its API and ChatGPT services. The issue occurred on December 5, from 7:45 to 7:50 UTC+8, causing temporary disruptions. By 8:08 UTC+8, the performance had stabilized, and OpenAI confirmed that they are actively monitoring the situation to prevent any recurrence.The latest incident was reported at 8:29 UTC+8 on the same day, with OpenAI's team working diligently to resolve the matter. The company is committed to ensuring the stability and reliability of its services and is investigating the root cause of these disruptions. OpenAI continues to keep users informed about the status of their services and any potential impacts on performance.

According to Odaily, OpenAI has announced a 12-day event starting this Thursday, during which the company will unveil a series of new features. The announcement was made via a post on X, stating, "12 days, 12 live streams, a bunch of new things big and small, OpenAI's 12-day event period starts tomorrow."OpenAI's CEO, Sam Altman, elaborated that beginning at 10 a.m. Pacific Time on Thursday (2 a.m. UTC+8 on Friday), the company will host a live stream each weekday to showcase a new product or sample. These presentations will include some major new products as well as smaller "gifts." The initiative reflects OpenAI's commitment to continuous innovation and engagement with its audience, offering insights into the company's latest developments and technological advancements. 

According to Decrypt, OpenAI CEO Sam Altman has criticized the New York Times for its lawsuit against the AI developer, accusing them of copyright infringement. The lawsuit, filed in December, claims that OpenAI and Microsoft used New York Times articles to train AI models without proper licensing. Altman expressed his views during an interview with New York Times journalist Andrew Ross Sorkin at the DealBook Summit in New York City.Altman refrained from discussing specifics but suggested that the New York Times is on the wrong side of history regarding AI's role in the media industry. He emphasized the need for a fair system to compensate creators for the use of their work, proposing an opt-in model where creators could earn micropayments when their content is used to generate AI responses.The New York Times alleges that OpenAI prioritized their content when developing large language models (LLMs) like ChatGPT, which are trained on extensive datasets to understand language patterns. OpenAI, however, disputes these claims, arguing that the Times manipulated prompts to make ChatGPT produce specific responses. The AI company contends that their models do not typically behave as the Times suggests, implying that the newspaper either instructed the model to regurgitate content or selectively chose examples.The lawsuit is part of a broader wave of legal actions against OpenAI, with other plaintiffs including authors George R.R. Martin, John Grisham, and comedian Sarah Silverman. Recently, a federal judge granted a motion by the Authors Guild to compel OpenAI to produce communications from employees who used social media for work purposes.Altman advocates for new economic models to support creators, suggesting that discussions on fair use need to evolve. He believes that creators should have opportunities for new revenue streams, aligning with a right-to-learn approach that balances innovation with fair compensation.

According to PANews, OpenAI's popular chatbot, ChatGPT, has reached a significant milestone by surpassing 300 million weekly active users. This information was disclosed by the company's CEO, Sam Altman, during The New York Times Dealbook Summit on Wednesday.This achievement comes just over three months after the chatbot reached 200 million weekly active users and a year after it hit the 100 million mark. Additionally, more than 10 million users have subscribed to the premium version of ChatGPT, highlighting its growing popularity and user engagement.

According to PANews, the United States Federal Trade Commission (FTC) has initiated an investigation into Microsoft's artificial intelligence software sales and its dealings with OpenAI. This development comes as regulatory bodies increasingly scrutinize major technology companies and their business practices. The investigation aims to assess whether Microsoft's activities in the AI sector, particularly its transactions with OpenAI, comply with existing trade regulations. The FTC's focus on these dealings highlights the growing concern over the influence and reach of AI technologies in the market. As AI continues to evolve and integrate into various sectors, regulatory bodies are keen to ensure that these advancements do not compromise competitive practices or consumer rights. Microsoft's collaboration with OpenAI has been a significant aspect of its AI strategy, with the company investing heavily in AI research and development. The outcome of this investigation could have implications for Microsoft's future AI initiatives and its position in the technology industry. The FTC's actions reflect a broader trend of increased regulatory oversight on tech giants, as authorities seek to balance innovation with fair market practices.

According to Cointelegraph, top lawmakers in the United States have introduced legislation aimed at understanding the effects of artificial intelligence on the financial services and housing sectors. Congresswoman Maxine Waters, alongside House Financial Services Committee Chair Patrick McHenry, has put forward a bill that mandates federal financial regulators to examine both the current and potential impacts of AI in these industries. This initiative is part of a broader resolution acknowledging the growing role of AI in finance and housing.The proposed "AI Act of 2024," sponsored by Waters, requires key regulatory bodies such as the Federal Reserve and the Federal Deposit Insurance Corporation to assess how banks are utilizing AI technologies to combat money laundering, cybercrime, and fraud. Waters emphasized the significant influence AI already has on mortgage lending and credit scoring, highlighting the necessity for a comprehensive regulatory framework to oversee AI reporting. McHenry's resolution further notes the use of AI in market surveillance and tenant screening, underscoring the technology's expanding footprint in these sectors.McHenry described the legislative efforts as a crucial step towards harnessing the benefits of AI for consumers, businesses, and regulators. He also suggested that the House Financial Services Committee should explore potential reforms to privacy laws in response to the increasing reliance on AI-driven data usage. McHenry expressed a desire for the United States to maintain its leadership in AI development and application.These legislative measures are an extension of the House Committee’s Bipartisan AI Working Group, which was established on January 11, 2024. The group includes Republican members such as French Hill, Young Kim, Mike Flood, Zach Nunn, and Erin Houchin, as well as Democrat members like Stephen Lynch, Sylvia Garcia, Sean Casten, Ayanna Pressley, and Brittany Pettersen. The formation of this group followed an executive order by US President Joe Biden on October 30, aimed at ensuring the safe, secure, and trustworthy development and use of artificial intelligence.

According to Foresight News, OpenAI is exploring the introduction of advertising into its artificial intelligence products as a means to generate new revenue streams. This development comes as the company, known for developing ChatGPT, seeks to diversify its income sources. In an interview with the Financial Times, OpenAI's Chief Financial Officer, Sarah Friar, revealed that the company, valued at $150 billion, is contemplating an advertising model. Friar emphasized that OpenAI intends to be cautious about when and where they implement advertising. The San Francisco-based company recently secured $6.6 billion in new funding in October. As part of its strategy to incorporate advertising, OpenAI is reportedly recruiting talent from major tech companies like Meta and Google. This recruitment effort has been confirmed by multiple sources and an analysis of LinkedIn profiles conducted by the Financial Times. The move to consider advertising reflects OpenAI's broader strategy to enhance its financial sustainability while continuing to innovate in the AI sector.