IBL News

Category: Top News

  • U.S. Colleges Advise International Students to Return Ahead of Trump Inauguration

    U.S. Colleges Advise International Students to Return Ahead of Trump Inauguration

    IBL News | New York

    Harvard, MIT, Cornell, UPenn, Wesleyan, and U.S.C. Cornell are among several American colleges advising their international students to return to campus before President-elect Donald J. Trump is inaugurated on January 20, The New York Times reported.

    During his last administration, President Trump imposed restrictions on entry to the U.S. from seven Muslim countries, a policy that stranded thousands of foreign students.

    “A travel ban is likely to go into effect soon after inauguration,” Cornell’s Office of Global Learning warned students on its website last month, advising them to return to the United States before spring-semester classes on Jan. 21.

    “The ban is likely to include citizens of the countries targeted in the first Trump administration: Kyrgyzstan, Nigeria, Myanmar, Sudan, Tanzania, Iran, Libya, North Korea, Syria, Venezuela, Yemen, and Somalia. New countries could be added to this list, particularly China and India,” said this institution.

    Colleges are also warning all students to prepare for possible delays at the border and in the processing of paperwork.

    However, the policies of the new Trump Administration remain uncertain.

    Over 1.1 million students from outside the United States were enrolled in American colleges and universities in the 2023-24 academic year. India, China, and South Korea have the highest number of students.

  • A New Way to Access ChatGPT: Through an Old-School 1-800 Number

    A New Way to Access ChatGPT: Through an Old-School 1-800 Number

    IBL News | New York

    OpenAI launched a new way to interact with its ChatGPT chatbot: through an old-school 1-800 number: 1-800-CHATGPT.

    By dialing the U.S. number (1-800-242-8478) from any phone device or messaging it via WhatsApp, users can access an “easy, convenient, and low-cost way to try it out through familiar channels” without needing an account.

    At first, the company said callers would get 15 minutes free per month.

    The WhatsApp version runs on a lighter model with daily usage caps, offering potential future upgrades like image analysis.

    The announcement on Wednesday is intended to expand OpenAI’s reach beyond the tech-savvy crowd.

    It follows a series of updates as part of its 12-day release event. The most notable announcement was the official rollout of Sora.

    In October, the company debuted a search feature within ChatGPT that positions it to better compete with Google, Microsoft’s Bing, and Perplexity.

    It’s all part of a growth plan for OpenAI to battle Amazon-backed Anthropic, Elon Musk’s xAI, Google’s Gemini, Meta, Microsoft, and Amazon in the generative AI market, predicted to top $1 trillion in revenue within a decade.

    OpenAI closed its latest funding round in October at a valuation of $157 billion. The company also received a $4 billion revolving line of credit, bringing its total liquidity to over $10 billion.

     

  • Harvard Releases a Dataset of One Million Books that Can Be Used to Train LLMs

    Harvard Releases a Dataset of One Million Books that Can Be Used to Train LLMs

    IBL News | New York

    Harvard University announced the release of a high-quality dataset of one million public-domain books this month.

    This dataset includes books scanned as part of the Google Books project, which are no longer protected by copyright. It spans genres, decades, and languages, with classics from Shakespeare, Charles Dickens, and Dante.

    It was created by Harvard’s newly formed Institutional Data Initiative with funding from both Microsoft and OpenAI.

    Anyone can use this dataset to train LLM and other AI tools.

    In addition to the trove of books, the Institutional Data Initiative is also working with the Boston Public Library to scan millions of articles from different newspapers now in the public domain, and it says it’s open to forming similar collaborations.

    Other new public-domain datasets, like Common Corpus, are also available on the open-source AI platform Hugging Face.

    It contains an estimated 3 to 4 million books and periodical collections.

    It was rolled out this year by the French AI startup Pleis, and the French Ministry of Culture backs it.

    Another one is called Source.Plus. It contains public-domain images from Wikimedia Commons as well as a variety of museums and archives.

    Several significant cultural institutions have long made their archives accessible to the public as standalone projects, like the Metropolitan Museum of Art.

    Ed Newton-Rex, a former executive at Stability AI who now runs a nonprofit that certifies ethically-trained AI tools, says the rise of these datasets shows that there’s no need to steal copyrighted materials to build high-performing and quality AI models.

    OpenAI previously told lawmakers in the United Kingdom that it would be “impossible” to create products like ChatGPT without using copyrighted works.

    “Large public domain datasets like these further demolish the ‘necessity defense’ some AI companies use to justify scraping copyrighted work to train their models,” Newton-Rex said to Wired.

  • Google Introduced ‘Agentspace’ to Bring Together AI Agents and AI-Search for Enterprise

    Google Introduced ‘Agentspace’ to Bring Together AI Agents and AI-Search for Enterprise

    IBL News | New York

    Google Cloud introduced Agentspace this month, a tool that includes agents that combine Gemini’s advanced reasoning, Google’s search, and enterprise data.

    • NotebookLM Plus is available as an out-of-the-box agent in Agentspace. It offers employees the features of NotebookLM and its podcast-like audio summaries, now enhancing it with security and privacy features for work.
    • Google Agentspace provides a single, company-branded multimodal search for the entire organization with the ability to answer complex questions, make proactive suggestions, and take action.
    • It also provides a single space for all custom expert AI agents used within a company to help employees conduct better research, draft content quickly, and automate repetitive tasks.

    • Details of Agentspace are on the Google Cloud blog.

     

  • Silicon Valley Billionaires Go on a Pilgrimage to Trump’s Mar-a-Lago to Bend the Knee

    Silicon Valley Billionaires Go on a Pilgrimage to Trump’s Mar-a-Lago to Bend the Knee

    IBL News | New York

    Silicon Valley billionaires from Meta, Amazon, Apple, Google, and OpenAI have traveled this month to President-elect Trump’s Mar-a-Lago residence in Florida to bend the knee while promising to support his inaugural committee with $1 million donations.

    The New York Times reported that tech leaders who traveled to hobnob with Mr. Trump face-to-face included Sundar Pichai, Google’s chief executive, and Sergey Brin, a Google founder, who dined with Mr. Trump on Thursday.

    Apple’s chief executive, Tim Cook, shared a meal with Mr. Trump on Friday. And Jeff Bezos, the founder of Amazon, planned to meet with Mr. Trump in the next few days.

    Also, the leaders of Meta, Amazon, and OpenAI’s Sam Altman promised to support President-elect Donald J. Trump’s inaugural committee with seven-figure checks over the past week, often accompanied by a pilgrimage to Mar-a-Lago to bend the knee.

    Tech companies and their top executives, reluctant for years to deal with Donald Trump, acknowledged the reality of getting business done in his new Washington.

    “President Trump will lead our country into the age of A.I., and I am eager to support his efforts to ensure America stays ahead,” Sam Altman, who personally wrote a $1 million donation to Mr. Trump’s inaugural fund, said in a statement.

    Other tech leaders have also praised Mr. Trump. Marc Benioff, the CEO of Salesforce and the owner of Time Magazine, posted on Thursday that it was “a time of great promise for our nation” after Time awarded Mr. Trump its coveted “Person of the Year” designation.

    Marc Benioff and Sam Altman have been among the most politically active Democratic tech donors during Mr. Trump’s first term.

  • X Added a New Image Generator Named Aurora

    X Added a New Image Generator Named Aurora

    IBL News | New York

    This month, the AI assistant on X released a new image generator called Aurora. xAI, Musk’s AI startup, developed it, and it is accessible through the Grok tab on X’s mobile apps and the web.

    “This is our internal image generation system,” Musk wrote Saturday in a post on X. “Still in beta, but it will improve fast.”

    Without complaint, it can generate images of public and copyrighted figures, like Mickey Mouse.

    TechCrunch said that the model stopped short of nudes, but graphic content, like “an image of a bloodied Trump,” wasn’t off limits.

    The company didn’t reveal whether xAI trained Aurora itself, built on top of an existing image generator, or, as was the case with xAI’s first image generator, Flux, collaborated with a third party.

    The release of Aurora comes after X made Grok free for all users, who can send up to 10 messages to Grok every two hours and generate up to 3 images per day.

    In other X and xAI news this month, xAI closed a $6 billion funding round.

  • OpenAI Blamed Its Three Hour Outage on a “New Telemetry Service”

    OpenAI Blamed Its Three Hour Outage on a “New Telemetry Service”

    IBL News | New York

    OpenAI blamed the major disruption of its ChatGPT, Sora, and API on December 11 on a new telemetry service to collect Kubernetes that went awry. The company admitted that “it had fallen short of our own expectations.”

    “This event resulted from an internal change to roll out new telemetry across our fleet,” said the company.

    The outage of three hours, between 3:16 PM PST and 7:38 PM PST, was one of the most prolonged outages in its history.

    A security incident or a recent product launch didn’t cause the downtime. In a postmortem explanation, OpenAI explained, “The issue stemmed from a new telemetry service deployment that unintentionally overwhelmed the Kubernetes control plane, causing cascading failures across critical systems.”

    Kubernetes is an open-source program that helps manage containers or packages of apps and related files used to run software in isolated environments.

    “Our Kubernetes API servers became overwhelmed, taking down the Kubernetes control plane in most of our large Kubernetes clusters.”

    OpenAI said that it could detect the issue “a few minutes” before customers ultimately started seeing an impact but that it couldn’t quickly implement a fix because it had to work around the overwhelmed Kubernetes servers.

    “This was a confluence of multiple systems and processes failing simultaneously and interacting in unexpected ways,” the company wrote.

    OpenAI said that it’ll adopt several measures to prevent similar incidents from occurring in the future.

     

  • Amazon Announces Its Nova Models and Other Initiatives to Become an AI Powerhouse

    Amazon Announces Its Nova Models and Other Initiatives to Become an AI Powerhouse

    IBL News | New York

    Amazon/AWS announced in December a series of AI projects, with several generative models, a new AI computer chip Trainium2, plans for a new supercomputer, and big partnership deals, including its plans to double its investment in Anthropic to $8 billion.

    With these announcements, Amazon signaled its intent to go beyond being a cloud platform and aim to be an end-to-end AI powerhouse.

    Trainium2 chips are specifically designed for heavy computing demands and are intended to be a direct competitor to Nvidia and AMD.

    Rainier’s supercomputer will be developed to meet its AI training and computational needs. It will rival Elon Musk’s Cortex and Colossus AI supercomputer.

    To compete with ChatGPT and Gemini, Amazon launched six foundational large language models under its Nova umbrella.

    This new family of multimodal AI models, Nova, includes four text-generating models: Micro, Lite, Pro, and Premier.

    In addition, there’s an image-generation model, Nova Canvas, and a video-generating model, Nova Reel.

    Micro has a 128,000-token context window, processing up to around 100,000 words. Lite and Pro have 300,000-token context windows, up to around 225,000 words, 15,000 lines of computer code, or 30 minutes of footage.

    AWS said that in early 2025, certain Nova models’ context windows will expand to support over 2 million tokens.

    The Nova models are available in AWS Bedrock, Amazon’s AI development platform. There, they can be fine-tuned on text, images, and video and distilled for improved speed and efficiency.

    These models are optimized to work with proprietary systems and APIs, allowing for multiple orchestrated automatic steps—agent behavior.

    “In 2025, we will bring even more capable models while driving cost down through algorithmic and computing innovations. And we will continue to invest in the talent, technology, and infrastructure necessary to offer world-class models to our customers for years to come,” Amazon said.

  • Google Issues Its New Model, ‘Gemini 2.0 Flash’, Along With ‘Multimodal Live API’

    Google Issues Its New Model, ‘Gemini 2.0 Flash’, Along With ‘Multimodal Live API’

    IBL News | New York

    Google announced yesterday its next major model, Gemini 2.0 Flash, which includes new multimodal outputs and can natively generate images, audio, and text. 2.0 Flash can.

    It uses third-party apps and services, allowing it to tap into Google Search, execute code, and more.

    However, the audio and image generation capabilities are launching only for “early access partners,” while the production version of 2.0 Flash will land in January.

    In the meantime, Google is releasing an API, the Multimodal Live API, to help developers build apps with real-time audio and video streaming functionality.

    Google said that using the Multimodal Live API allows developers to create real-time, multimodal apps with audio and video inputs from cameras or screens.

    The API supports the integration of tools to accomplish tasks, and it can handle “natural conversation patterns” such as interruptions along the lines of OpenAI’s Realtime API.

    The Multimodal Live API was generally available as of yesterday.

    In addition, Google released Jules, an experimental AI-powered code agent using Gemini 2.0. for coding tasks with Python and Javascript. Jules creates comprehensive, multi-step plans to address issues, efficiently modifies multiple files, and even prepares to pull requests to land fixes into GitHub directly.

  • ChatGPT-4o’s Canvas, which Allows a Live Editing Workspace, Made Available to All Users

    ChatGPT-4o’s Canvas, which Allows a Live Editing Workspace, Made Available to All Users

    IBL News | New York

    OpenAI made its collaborative split-screen writing and coding interface, Canvas, available to all users this week. Integrated natively with GPT-4o, this tool has gained new features like Python and direct code execution within the interface, supporting real-time debugging and output visualization.

    Canvas allows users to trigger the interface through prompts rather than manual model selection.

    It features a split-screen layout with the chat on one side, a live editing workspace on the other, editing tools for writing (reading level, length adjustments), and advanced coding tools (code reviews, debugging).

    OpenAI introduced Canvas as an early beta to Plus and Teams users in October. This month, with the full rollout, all accounts now have access.