IBL News

Category: Top News

  • OpenAI’s GPT-5 Rollout Faced Backlash as Old Models Were Retired

    OpenAI’s GPT-5 Rollout Faced Backlash as Old Models Were Retired

    IBL News | New York

    OpenAI’s last upgrade to GPT-5 rollout faced backlash for retiring older models among users. OpenAI acknowledged that it underestimated users’ affection for the older GPTs, even if GPT-5 performs better in most ways.

    In response to the critics, OpenAI’s CEO, Sam Altman, announced that rate limits for ChatGPT Plus users would be doubled and would continue to use the 4o model.

    Previously, ChatGPT could tap into several different AI models, including GPT‑4o, o3, o4-mini, GPT‑4.1, and GPT‑4.5. But OpenAI has since replaced them with a family of GPT-5 models.

    On Thursday, OpenAI unveiled a new flagship AI model, GPT-5, and began sharing the technology with ChatGPT users worldwide.

    OpenAI executives called GPT-5 a “major upgrade” over their AI systems, saying the new technology was faster, more accurate, and less likely to hallucinate.

    “GPT-5 is the first time that it feels like talking to an expert in any topic — a Ph.D.-level expert,” said OpenAI’s CEO, Sam Altman.

    Also, it was the first time that OpenAI has used a reasoning model to power the free version of ChatGPT.

    Experts agreed on the fact that the technology feels more human than previous models.

    Sam Altman called the system a “significant step” along the path to the ultimate goal of the company and its rivals: artificial general intelligence, or AGI, a machine that can do anything the human brain can do.

    GPT-5’s launch arrives in a moment when OpenAI, which is not yet profitable, plans to raise $40 billion this year, while being on pace to generate revenues of $20 billion by the end of 2025.

  • OpenAI Introduces Its Flagship Model ‘GPT-5’, Making It the New Default in ChatGPT

    OpenAI Introduces Its Flagship Model ‘GPT-5’, Making It the New Default in ChatGPT

    IBL News | New York

    OpenAI unveiled its latest model, GPT-5 yesterday, making it available to all ChatGPT users, including those on the free tier, with usage limits varying by subscription level.

    It is also integrated into Microsoft 365 Copilot and available to developers through the OpenAI API.

    The company made GPT‑5 the new default in ChatGPT, replacing GPT‑4o, OpenAI o3, OpenAI o4-mini, GPT‑4.1, and GPT‑4.5 for signed-in users.

    GPT‑5 started to roll out today to all Plus, Pro, Team, and Free users, with access for Enterprise and Edu coming in one week. Pro, Plus, and Team users could also start coding with GPT‑5 in the Codex CLI⁠ by signing in with ChatGPT.

    Pro subscribers got unlimited access to GPT‑5, and access to GPT‑5 Pro, for complex tasks and a replacement of OpenAI o3‑pro. Plus users have significantly higher usage than free users.

    “For ChatGPT free-tier users, full reasoning capabilities may take a few days to fully roll out. Once free users reach their GPT‑5 usage limits, they will transition to GPT‑5 mini, a smaller, faster, and highly capable model,” said OpenAI.

    During a livestream presentation [watch in the video below], the company’s CEO, Sam Altman, described the model as “having a team of experts on call for whatever you want to know.” He defined it as “our smartest, fastest, most useful model yet, with built-in thinking that puts expert-level intelligence in everyone’s hands.”

    GPT‑5 provides more useful responses across math, science, finance, and law. It includes 128,000 max output tokens. The price per 1 million tokens for the API is $1.25.

    The tools supported by this model when using the Responses API are Web search, File search, Image generation, Code interpreter, and MCP.

    OpenAI’s executives who participated in the virtual presentation highlighted the coding capabilities of GPT-5, tackling “complex tasks end-to-end and delivering more readily usable code, better design, and is more effective at debugging.”

    GPT‑5 is OpenAI’s most advanced model for coding and agentic tasks. It produces high-quality code, generates front-end UI with minimal prompting, and shows improvements to personality, steerability, and executing long chains of tool calls.”

    GPT‑5 also showed ‘minimal’ reasoning and a ‘verbosity’ parameter in the API.

     

    • Pricing and Characteristics
    • Release Notes
    • Research Paper

  • Gemini Introduces Its Socratic-Style AI Assistant for Learners

    Gemini Introduces Its Socratic-Style AI Assistant for Learners

    IBL News | New York

    Google’s Gemini introduced yesterday its “education mode”, a Socratic-style AI companion tutor similar to OpenAI’s Study Mode and Anthropic’s Claude for Education, both recently launched.

    Like the other personalized assistants, “Guided Learning in Gemini” is designed to avoid immediately spitting out quick answers and ask probing, open-ended questions, adapt explanations to the learner’s level, walk them through step-by-step reasoning, and use videos, diagrams, visuals, and quizzes to reinforce concepts.

    Essentially, they all guide users through problems with Socratic prompts, scaffolded reasoning, and adaptive feedback across a range of subjects and skill levels, instead of just handing over the answer.

    These AI companies said that these AI assistants are built with input from educators, pedagogical experts, and learning scientists, alongside feedback from college students.

    “Guided Learning is designed to be a partner in teaching, built on the core principle that real learning is an active, constructive process. It encourages students to move beyond answers and develop their own thinking by guiding them with questions that foster critical thought. To make it simple to bring this approach into their classrooms, we created a dedicated link that educators can post directly in Google Classroom or share with students,” explained the search giant.

    However, independent experts argue that these mentors show many limitations, such as minimal persistent memory, early over-structuring, and a tendency to agree too quickly.

    In the educational area, Gemini is already getting a good response with LearnLM, a family of models fine-tuned for learning and grounded in educational research.

  • Anthropic Launches Claude for Education In AWS Marketplace

    Anthropic Launches Claude for Education In AWS Marketplace

    IBL News | New York

    Anthropic’s Claude for Education offering was made available in AWS Marketplace as a software-as-a-service solution this month.

    Claude for Enterprise and the Financial Analysis Solution were also released at the same space.
    AWS Marketplace’s main advantage is the streamlined procurement and billing process.

    According to the company, “Claude for Education equips every student with an adaptive study companion, faculty with an AI assistant for creating engaging teaching materials, and staff with an AI collaborator for tracking and analyzing student progress.”

    Claude for Education uses Socratic questioning to guide students toward answers rather than providing direct responses.

    It includes single sign-on (SSO), native integrations with GitHub, Google Workspace, and Canvas LTI (with Panopto and Wiley integrations coming soon), and custom integrations through the Model Context Protocol (MCP). The pre-built MCP integrations include Atlassian (Jira/Confluence), Zapier, Linear, and Asana.

    It adds a 200K token context window, primarily for analysis of complex academic materials and research tasks in a single conversation, enterprise-grade security, and compliance.

    For example, a research team can upload multiple academic papers, datasets, and their own notes into a single Claude conversation. Claude maintains full context across all documents, enabling comprehensive analysis and synthesis that would typically require hours of manual work.

    Antropic highlighted that its key use cases include:

    • Academic instruction and learning: Socratic questioning through Learning Mode
    • Faculty support: Course development and content creation
    • Research support: Literature review and data analysis assistance
    • Student success support: Progress tracking, early intervention strategies, and personalized learning paths

    ——

    • Claude for Education in AWS Marketplace

    • Kim Majerus’s keynote at the AWS Imagine: Education, State, and Local Government conference.

  • Canvas LMS Adds to Its Platform an Agentic Solution, ‘IgniteAI’

    Canvas LMS Adds to Its Platform an Agentic Solution, ‘IgniteAI’

    IBL News | New York

    Instructure, the maker of Canvas LMS, announced last month the launch of its native AI solution called IgniteAI.

    Powered by AWS Bedrock, IgniteAI is embedded within Canvas and Mastery to conduct tasks such as creating quizzes, generating rubrics, summarizing discussions, and aligning content to outcomes.

    It also leverages the Model Context Protocol (MCP) standard and extends the LTI framework. This way, Canvas LMS’ ecosystem of 1,100 edtech partners and LLMs like Anthropic and OpenAI can integrate their agentic AI solutions.

    IgniteAI emphasizes data protection compliance with COPPA, FERPA, and GDPR. In terms of accessibility, Canvas LMS and other products are achieving WCAG 2.2 AA compliance as part of a recent Voluntary Product Accessibility Test (VPAT).

    In addition, Instructure announced several updates to its product suite Canvas, featuring redesigned dashboards and modules, an improved mobile app, specialized STEM items, enhanced proctoring capabilities in New Quizzes, and new student portfolios that showcase diverse learning progress.

     

  • Satya Nadella Explains the 9,000-Employee Layoff While Microsoft Thrives

    Satya Nadella Explains the 9,000-Employee Layoff While Microsoft Thrives

    IBL News | New York

    Satya Nadella, Microsoft’s CEO, rationalized the 9,000-employee layoff through a 1,150-word memo that highlighted how the company is thriving in terms of market performance, strategic positioning, and growth, while the AI-based disruption is taking place in the software industry.

    Progress isn’t linear. It’s dynamic, sometimes dissonant, and always demanding. But it’s also a new opportunity for us to shape, lead through, and have greater impact than ever before.

    In a double corporate language, Nadella tries to reconcile two contradictory realities: How can a company be “more successful than ever” while still eliminating jobs?

    “This narrative framework captures the harsh reality that AI, in theory, will make companies more profitable while employing fewer people,” wrote San Francisco-based writer, photographer, and investor Om.

    Microsoft’s CEO [in the picture above] implies that the laid-off employees are not due to financial struggles, but rather because those workers didn’t align with the company’s AI-focused strategy, suggesting that some employees’ skills have become outdated.

    Rather than invest in retraining, the company opted to hire fewer workers with more relevant expertise.

    “The Microsoft memo portends the new reality of the technology industry. For years, the sector has been generous to its employees, offering unheard-of perks and placing a premium on skills such as software development. AI, however, inverts that relationship,” said Om.

  • Figma Makes Its AI Coding Tool ‘Make’ Available to All Users

    Figma Makes Its AI Coding Tool ‘Make’ Available to All Users

    IBL News | New York

    Figma announced that its AI coding tool, Figma Make, for building prototypes and apps, is now available to all users, with limitations depending on the subscription plan.

    Figma Make, still in beta, features the ability to include design references.

    Users can upload an image alongside the description of what they want to create, and elements like formatting and font style can also be adjusted using additional prompts.

    Figma has introduced a credit system for using the platform’s AI tools.

    According to the company, View, Collab, and Dev Seat users can use AI features with lower credit limits.

    In addition to Figma Make, the company is promoting two other AI features: Make and Edit Image and Boost Resolution. These products are transitioning from beta to general availability.

  • ChatGPT Introduced “Study Mode”, a New Way to Learn that Offers Step-By-Step Guidance

    ChatGPT Introduced “Study Mode”, a New Way to Learn that Offers Step-By-Step Guidance

    IBL News | New York

    OpenAI introduced “Study Mode” for ChatGPT, a new way to learn that offers step-by-step guidance instead of quick answers, designed to act more as an always-on tutor and less like a lookup tool. It aims to prevent—or at least discourage—students from taking homework shortcuts.

    The company defined this new feature as “a new learning experience that helps you work through problems step-by-step instead of just getting an answer.”

    This mode transforms the AI from an answer engine into a Socratic tutor, a pedagogical approach, developed after consulting with experts from over 40 institutions, that asks guiding questions to help students work through problems themselves instead of providing direct solutions.

    OpenAI states that it is currently partnering with learning experts from Stanford “to study and share how AI tools, including study mode, influence learning outcomes in areas like K-12 education.”

    The company aims to address educators’ concerns about academic integrity and cheating.

    The feature is available to logged-in users on Free, Plus, Pro, and Team, with availability in ChatGPT Edu in August.

    These are the key features, according to OpenAI:

    • Interactive prompts: Combines Socratic questioning, hints, and self-reflection prompts to guide understanding and promote active learning, instead of providing answers outright.

    • Scaffolded responses: Information is organized into easy-to-follow sections that highlight the key connections between topics, keeping information engaging with just the right amount of context and reducing overwhelm for complex topics.

    • Personalized support: Lessons are tailored to the right level for the user, based on questions that assess skill level and memory from previous chats.

    • Knowledge checks: Quizzes and open-ended questions, along with personalized feedback to track progress, support knowledge retention, and the ability to apply that knowledge in new contexts.

    • Flexibility: Easily toggle study mode on and off during a conversation, giving you the flexibility to adapt to your learning goals in each conversation.

    However, regardless of how engaging ChatGPT’s study mode becomes, it exists just a toggle click away from ChatGPT with direct answers. That could be quite hard to resist for many students.

    In terms of market value, OpenAI’s “Study Mode” intensified the race among tech giants, with Google, Microsoft, and Anthropic each competing to shape the future of education.

  • Brown University Will Spend $50M on State Programs to See Its Federal Funding Restored

    Brown University Will Spend $50M on State Programs to See Its Federal Funding Restored

    IBL News | New York

    Brown became the third Ivy League university in a month to reach an agreement with the Trump administration to see research funding restored.

    In April, the Trump administration blocked $510 million in federally sponsored medical and health sciences research funding from flowing to the school. In its 2024 fiscal year, Brown received about $184 million through federal grants and contracts.

    The new deal, announced yesterday, requires Brown University to spend $50 million in Rhode Island’s workforce development programs over a decade. It also requires the institution to comply with the White House’s vision on transgender athletes, anti-semitism, and merit-based admission policies.

    Per the agreement, Brown will dictate its curriculum and the content of academic speech.

    In an open letter on Wednesday, Brown’s President, Christina H. Paxson, said the agreement “preserves the integrity of Brown’s academic foundation, and it enables us as a community to move forward after a period of considerable uncertainty.”

    However, the Trump administration depicted the deal as an ideological victory. The education secretary, Linda McMahon, said the deal would be part of a “lasting legacy of the Trump administration, one that will benefit students and American society for generations to come.”

    “The Trump administration is successfully reversing the decades-long woke-capture of our nation’s higher education institutions,” Ms. McMahon stated.

    Observers, including the student community and media outlets like The New York Times, saw the deal as capitulation to the Trump administration.

    This month, the U.S. government has also reached agreements with the University of Pennsylvania and Columbia University.

    White House officials are still negotiating with Harvard University and representatives of other schools.

    On the other hand, the Trump Administration accused Duke University of “systemic racial discrimination” in its health care system and froze $108 million in federal funds.

    Amid this budget crunch, the institution is weighing layoffs alongside about $350 million in cuts, amounting to roughly 10 percent of its budget.

  • Learners Use AI to Redesigning Instructional Experiences in Real-Time

    Learners Use AI to Redesigning Instructional Experiences in Real-Time

    IBL News | New York

    Learners are increasingly becoming instructional designers, identifying gaps in the learning experience and utilizing AI to address these gaps in pedagogical practices.

    These instructional design gaps can be inadequate or inaccessible content, poor practice opportunities, deficient scaffolding, and missing emotional support, among others.

    For example, one student said, “I use ChatGPT as a study guide to explain stuff that the course glosses over, which I then add to my notes. This helps me reinforce what I’m learning, and it’s been hella useful so far.”

    Additionally, learners asking AI to “act as my professor and grade this draft” or “create a practice quiz, asking me each question one by one” reveal that instructional assessment and practice systems are failing to support learning.

    Therefore, experts suggest that instructional designers should closely study how learners interact with AI and adapt their role accordingly, evolving from content creators to learning ecosystem architects.

    OpenAI’s “Top 20 Chats for Finals” data reveal how learners worldwide actually engage with learning content.

    In a May 2025 post, OpenAI disclosed how many learners are using AI to enhance their learning process.

    These are some prompts:

      • “I want to learn by teaching. Ask me questions about [topic] so I can practice explaining the core concepts to you.”
      • “Identify and share the most important 20% of learnings from this topic that will help me understand 80% of it.”
      • “Create a practice quiz for me based on the material. Ask me each question one by one.”
      • “I’m not feeling it today. Help me understand this lecture, knowing that’s how I feel.”
      • “Motivate me.”
      • “Can you take the following slides and help me learn the content in a faster and more interesting way?”
      • “Decode this dense passage into language I can understand.”
      • “Create a game to help me [learning goals]
      • “Give me a step-by-step guide to help me finish [project]. Make the steps as small and achievable as possible.”
      • “Look for any rules and requirements in this assignment and make a checklist that’s easy to understand.”
      • “Act as my public speaking coach and give me feedback to help me improve.”
      • “I want to pressure test my thesis before I keep writing. Suggest the existing opposing viewpoints and any flaws in my logic.”
      • I want to consider multiple perspectives. Find three experts with different points of view and compare their opinions.

    Dr. Philippa Hadman, an expert and researcher on education, wrote,

    “The most successful instructional designers of 2025 and beyond won’t be those who resist AI or those who blindly embrace it. The winners will be those who study what learner AI behaviour teaches us about effective learning design and who use those insights to create more responsive, human-centered learning ecosystems.”

    “The future of instructional design is about learning from what learners create when they have access to responsive, personalized, and emotionally intelligent learning support.”