83
OpenAI Unveils ‘Operator’ AI to Automate Web Tasks
Date: January 23, 2025Source: PureAIOpenAI has launched 'Operator,' an AI agent designed to perform web-based tasks autonomously using its own browser. Currently available as a research preview for Pro users in the U.S., Operator aims to automate everyday activities such as filling out forms, ordering groceries, and creating memes. It can interact with the web like a human, using a browser to click, type, and scroll, and is powered by OpenAI’s new Computer-Using Agent (CUA) model.Link: https://pureai.com/Articles/2025/01/23/OpenAI-Unveils-Operator-AI.aspxPope warns Davos summit that AI could worsen ‘crisis of truth’
Date: January 23, 2025Source: National Catholic ReporterPope Francis warned the World Economic Forum in Davos that artificial intelligence (AI) could exacerbate the 'crisis of truth' and urged leaders to ensure AI serves human dignity and does not worsen inequality.Link: https://www.ncronline.org/vatican/vatican-news/ai-human-creation-must-be-controlled-humans-pope-saysExclusive: ByteDance plans $20 billion capex in 2025, mostly on AI, sources say
Date: January 23, 2025Source: ReutersByteDance, the Chinese owner of TikTok, has earmarked over 150 billion yuan ($20.64 billion) in capital expenditure for this year, much of which will be centred on artificial intelligence. The spending will help ByteDance defend its AI lead at home and reinforce AI offerings abroad.Link: https://www.marketscreener.com/quote/stock/NVIDIA-CORPORATION-57355629/news/Exclusive-ByteDance-plans-20-billion-capex-in-2025-mostly-on-AI-sources-say-48847214/OpenAI tells India court ChatGPT data removal will breach US legal obligations
Date: January 23, 2025Source: ReutersOpenAI has informed the Delhi High Court that deleting training data for ChatGPT would violate its legal obligations in the United States. The company argued that Indian courts lack jurisdiction over the copyright breach case filed by ANI, as OpenAI has no physical presence in India. ANI seeks the deletion of its data stored in ChatGPT's system, which it claims was used without consent.Link: https://pressinsider.com/business/openai-tells-court-scrapping-chatgpt-data-will-breach-us-legal-obligations-report/Microsoft to train 1 million South Africans on AI skills
Date: January 23, 2025Source: Dunya NewsMicrosoft aims to provide 1 million people in South Africa with artificial intelligence (AI) and cybersecurity training opportunities by 2026. The initiative targets a broad audience, including companies, government entities, and youth. Microsoft's President for Africa, Lillian Barnard, emphasized the importance of equipping youth with future-ready skills to compete globally.Link: https://dunyanews.tv/en/Technology/864148-microsoft-to-train-1-million-south-africans-on-ai-skillsPope decries ‘crisis of truth’ in AI message to Davos forum
Date: January 23, 2025Source: Global Banking and Finance ReviewPope Francis urged leaders at the World Economic Forum in Davos to keep close oversight of AI, warning it can exacerbate a growing 'crisis of truth'. He praised AI's abilities but raised critical concerns about humanity's future. The pope emphasized the need for diligence and vigilance in AI development, highlighting its potential to worsen inequalities and conflicts.Link: https://www.globalbankingandfinance.com/DAVOS-MEETING-POPE-91ce7e39-fa8d-4683-ada6-2ca6d0780afaBreakthroughs in AI helping to design life-saving drugs faster
Date: January 23, 2025Source: Lead StoryA team of scientists at the University of Toronto has successfully implemented artificial intelligence (AI) tools to speed up the process of discovering life-saving drugs. This breakthrough is expected to significantly reduce the time and cost associated with drug development.Link: https://www.leadstory.com/v/breakthroughs-in-ai-helping-to-design-life-saving-drugs-faster-20251231Perplexity Adds AI Assistant That Calls Other Apps and Performs Tasks
Date: January 24, 2025Source: PYMNTSPerplexity AI has launched an assistant within its Android app that can help users with tasks like booking dinner, finding a forgotten song, calling a ride, drafting emails, and setting reminders. The new Perplexity Assistant uses reasoning, search, and apps to assist with daily tasks.Link: https://www.pymnts.com/artificial-intelligence-2/2025/perplexity-adds-ai-assistant-that-calls-other-apps-and-performs-tasks/AI isn’t very good at history, new paper finds
Date: January 19, 2025Source: TechCrunchA new benchmark, Hist-LLM, tested three top large language models (LLMs) on historical questions. The results showed that even the best-performing LLM, GPT-4 Turbo, achieved only about 46% accuracy, not much higher than random guessing. The researchers found that LLMs struggle with nuanced historical inquiries and tend to extrapolate from prominent historical data, leading to inaccuracies.Link: https://techcrunch.com/2025/01/19/ai-isnt-very-good-at-history-new-paper-finds/Meta’s Yann LeCun predicts ‘new paradigm of AI architectures’ within 5 years and ‘decade of robotics’
Date: January 23, 2025Source: TechJuiceMeta's Chief AI Scientist Yann LeCun predicts a new paradigm in AI architectures within five years. He criticizes current generative AI and large language models (LLMs) for their limitations, including insufficient comprehension of the physical world, lack of enduring memory, deficiency in reasoning abilities, and lack of advanced planning capabilities. LeCun envisions future AI systems that can understand the physical world, remember interactions, and reason. He also predicts a decade of robotics, with robots becoming pivotal in practical domains like law and medicine.Link: https://www.techjuice.pk/predictions-for-ai-and-robotics-big-changes-ahead-in-the-next-5-years/