Today’s AI news you might have overlooked: First Hindi LLM OpenHathi, India’s AI contribution praised by French Minister, and more.
Today, on December 13, there were significant developments in the field of artificial intelligence (AI), particularly in India. Sarvam AI, an Indian startup, has introduced OpenHathi-Hi-v0.1, the first large language model (LLM) in Hindi. Additionally, at the GPAI Summit, Jean-Noel Barrot, the French Minister Delegate for Digital Affairs, highlighted India’s prominent position in the field of AI. These are just a few highlights from today’s AI news. Let’s delve deeper into the details.
OpenHathi LLM was released
Indian AI startup Sarvam AI has launched OpenHathi-Hi-v0.1, the first Hindi large language model (LLM) in the OpenHathi series, reports Moneycontrols. It is the first Hindi language LLM to be built. The model developed for Meta AI’s Llama2-7B architecture claims to match the performance of GPT-3.5 for Indian languages. The AI model utilizes the 48,000 tokens extension of Llama2-7B’s tokenizer and undergoes a two-stage training process that includes embedding alignment in the first stage and bilingual language modeling in the second stage.
French minister talks about India’s role in artificial intelligence
French Minister of Digital Affairs Jean-Noel Barrot highlighted India’s significant role in artificial intelligence (AI) as host of the GPAI summit, reported ANI. On his second official visit to India, Barrot expressed his intention to strengthen bilateral cooperation in the field of digital technology. He highlighted the continued cooperation between India and France and stated that both countries share common values and are committed to advancing their relations and global agenda. In addition, Barrot plans to attend the GPAI Summit Ministerial Council in Delhi.
“India is playing a leading role in AI by hosting this summit by sharing a global partnership on AI next year. A lot has been achieved during India’s G20 chairmanship, so we will continue to build our relationship and global agenda together as we are well aligned and share the same values,” ANI quoted him as saying.
Japanese minister calls for safe and sustainable artificial intelligence
Japan’s Vice Minister for Policy Coordination Hiroshi Yoshida has called for the creation of a safe, secure and sustainable AI ecosystem, according to a report by ANI. In Wednesday’s talks, he particularly wanted India’s active participation in achieving this goal. Highlighting the importance of responsible AI, Yoshida highlighted Prime Minister Narendra Modi’s keen interest in AI policy and encouraged India to actively participate in shaping a responsible and safe AI environment.
“Artificial intelligence is a very important issue all over the world and it is changing our lives and our society and we need to work to reduce AI risk… GPAI is very important and we also discussed at the G7 and made a guiding principle for all AI actors regarding AI through the Hiroshima AI process” , Yoshida said while talking to ANI.
Meta ignored legal warnings, used illegally produced books to train AI, lawsuit claims
Meta Platforms, the parent company of Facebook and Instagram, received legal warnings about the potential legal risks of using thousands of pirated books to train its AI models, according to a Reuters report. However, according to a recent copyright infringement lawsuit, the company continued this practice. The lawsuit, originally filed by comedian Sarah Silverman, Pulitzer Prize winner Michael Chabon and other well-known authors, alleges that Meta used their works without permission to train its AI model, Llama. A California judge recently dismissed parts of Silverman’s lawsuit, but said the authors would be allowed to amend their claims.
NYT is hiring a news agency’s AI director
The New York Times has named Quartz founder Zach Seward as editorial director of artificial intelligence initiatives, reflecting a growing trend among media organizations exploring the use of artificial intelligence in newsrooms, reports Axios. In this new role, Seward will work with news agency leaders to define principles for the responsible use of generative AI and consider ethical considerations to maintain public trust as the industry continues to experiment with AI technology.
