Meta has developed AI models that can recognize and produce speech for over 1,000 languages, a tenfold increase from what is currently available. The company is releasing the models via GitHub to help developers build new speech applications in different languages. Meta researchers retrained an existing AI model on two new data sets containing audio recordings of the New Testament Bible and its corresponding text taken from the internet in 1,107 languages and unlabeled New Testament audio recordings in 3,809 languages. The team processed the speech audio and text data to improve quality before running an algorithm designed to align audio recordings with accompanying text. While impressive, there are concerns about using religious texts to train AI models due to potential bias and misrepresentations.

Meta Platforms, the owner of Facebook, has been fined $1.3 billion by European Union regulators for illegally sending user information to the US. The fine is a record privacy penalty for the bloc and raises pressure on the US government to complete a deal that would allow Meta and other multinational companies to continue sending such data stateside. The ruling also orders Meta to stop transferring information about Facebook's European users to the US and delete any data already sent within six months unless a new trans-Atlantic agreement is reached before then. Tech companies have faced increasing regulatory scrutiny over European privacy concerns since 2020 when EU courts overturned a previous data-sharing deal between the US and EU.

Meta, the social media giant, is reportedly launching a decentralized Twitter alternative in late June. The new app, which is yet to be named, will be compatible with Mastodon. The microblogging platform will have a character limit of up to 500 characters, similar to Twitter, and will offer features such as the ability to "tune out the noise" and talk directly with your audience. Meta is also focusing on other technology developments, including AI and the metaverse, as well as new hardware and infrastructure projects. The launch of the new app is part of Meta's bid to reinvent itself and offer a separate space for creators and public figures to share timely updates about their interests.

Facebook's parent company, Meta, is reportedly in early talks with augmented reality start-up Magic Leap to create a multi-year agreement that could include intellectual property licensing and contract manufacturing in North America for AR products.

Recent advancements in artificial intelligence (AI) are boosting optimism about the productivity of businesses in the future and contributing to a surge in the stock market.

Meta, the parent company of Facebook, has announced that it is investing heavily in artificial intelligence (AI). The company is building a new supercomputer called the RSC (Research SuperCluster) that will be used to train and run AI models. Meta is also developing two custom chips for AI: the MTIA (Meta AI Training Accelerator) and the MSVP (Meta AI Scaling Accelerator). The MTIA is designed to train AI models, while the MSVP is designed to run AI models. Meta says that the RSC and the custom chips will help it to develop new AI-powered features for its products, such as better speech recognition and translation. The company is also investing in AI research and development, and it has created a new AI research center in Paris. Meta's investment in AI is a sign of the company's commitment to the technology. AI is seen as a key to future growth for Meta, and the company is betting big on it.

Facebook, now known as Meta Platforms Inc, began its journey into the metaverse two years ago. The company has since been pushing towards Artificial Intelligence (AI) through various products and endorsements, leading to speculation that it may be pivoting towards AI. Meta has been researching AI technology, including real-time translation of languages, using AI for children's drawing animation, and decoding speech through imagery brain activity. While Meta has launched AI-inspired projects such as Galatica and BlenderBot 3, they have failed to gain traction. Despite this, Zuckerberg maintains that AI is important to the Metaverse and has led to good results in all of Meta's businesses and apps.

Meta has released an open-source AI model that combines six types of data, including text, audio, visual data, temperature, and movement readings. The model is called ImageBind and is a research project with no immediate consumer or practical applications. The core concept of the research is linking together multiple types of data into a single multidimensional index. This idea underpins the recent boom in generative AI. Meta's ImageBind model is the first to combine six types of data into a single embedding space. The six types of data included in the model are visual, thermal, text, audio, depth information, and movement readings generated by an inertial measuring unit. Future AI systems will be able to cross-reference this data in the same way that current AI systems do for text inputs. The research brings machines one step closer to humans' ability to learn simultaneously, holistically, and directly from many different forms of information. Meta is open-sourcing the underlying model, which is an increasingly scrutinized practice in the world of AI. The immediate applications of research like this will be much more limited, but it shows how future versions of the system could incorporate other streams of data.

A growing number of malware creators are now taking advantage of the significant interest in Chatgpt to lure victims, Facebook owner Meta has noticed.

On Wednesday, the Federal Trade Commission proposed sweeping changes to how Meta operates, accusing the company of violating a suite of child privacy protections, including the Children’s Online Privacy Protection Act (COPPA). The FTC's proposed changes would ban Meta, Facebook, and the rest of the company's properties from monetizing the data of children under 18 years old. The FTC's proposal is just the first step in a process to heighten Meta's privacy and security practices. The FTC has asked Meta to respond within 30 days to its findings and proposal. After receiving Meta's response, the agency would then vote on whether to alter or adopt the proposed rules.