Latest news with #AI-guided
Forbes
24-07-2025
- Business
- Forbes
Teachers Get A New Assistant: Instructure Drops AI Into Canvas
AI in education Instructure and OpenAI have announced a new partnership to bring LLM-powered AI technology into Canvas, one of the most widely used learning platforms in education. The collaboration introduces IgniteAI, a built-in set of generative AI tools that will be released to Canvas users in stages over the coming year. Where AI is Adding Value in Canvas A key piece of the IgniteAI rollout is a new assignment builder that lets educators create AI-guided tasks. Teachers can write learning goals and sample prompts, set up how the chatbot will interact with students, and define how outcomes should be evaluated. At the same time, Canvas's grading system, analytics tools, and content creation features get new automation support, from faster feedback to AI-generated rubrics. Teachers stay in full control of how the AI behaves. They can customize each prompt and review all chatbot responses. Meanwhile, students get a chance to have focused conversations with the AI inside Canvas, working through ideas at their own pace. All chats are visible to the instructor, and the company says student data stays local and is not shared with OpenAI. The system also tracks each student's interaction. When learners show understanding or make progress, those moments are captured and added to the Gradebook. That lets teachers see not just the end result, but how a student arrived there. Repetitive tasks such as rewriting rubrics, responding to common requests and drafting feedback are handled by the system, allowing instructors to focus on discussion, coaching, and more complex teaching. "We're committed to delivering next-generation LMS technologies designed with an open ecosystem that empowers educators and learners to adapt and thrive in a rapidly changing world," said Steve Daly, CEO of Instructure. "This collaboration with OpenAI showcases our ambitious vision: creating a future-ready ecosystem that fosters meaningful learning and achievement at every stage of education.' Opportunities and Tradeoffs Daly says this partnership will free up time for educators and give students a more flexible way to engage with lessons. Leah Belsky, who oversees education strategy at OpenAI, describes the tools as a way to offer 'more personalized and connected learning experiences,' without removing human oversight. Schools are already moving quickly. Surveys show education leading all sectors in generative-AI adoption. Early feedback from pilots suggests students feel more confident when they can test ideas in a private chat, and some classroom studies point to modest gains in test scores among students using AI for practice. Still, the tools raise concerns. Nearly half of faculty respondents in recent polls say they worry about bias in model outputs. A similar number cite data privacy as a top issue. Those who work on academic integrity expect new forms of cheating to emerge. Others warn that expensive AI licenses could deepen gaps between well-funded and under-resourced schools. And until teachers are fully trained on how to use the tools, confusion and uneven results are likely. A university survey from May 2025 confirmed many of these fears among students. Respondents cited grading fairness, misuse of AI for shortcuts, and the risk of over-relying on automated suggestions as top concerns. Faculty echoed those points. They questioned whether AI nudges weaker writers toward overly similar phrasing and whether automated grading could undermine trust. To reduce those risks, campuses are already setting up review boards, bias checks, and clear opt-out options. Instructure, for its part, says that all student data stays within the institution, and that OpenAI has no access to individual records. Privacy teams are expected to monitor that closely. Where This Leads Canvas is now placing AI tools where teaching already happens—in assignments, discussions, and grading workflows. The chatbot becomes part of the lesson, not just an external add-on. If the systems work as intended, teachers could gain clearer feedback and students could move beyond generic answers into more thoughtful, process-based work. If the technology fails to live up to that promise, trust may erode. Either way, AI is no longer sitting outside the classroom door. It's embedded, logged, and learning alongside everyone else.
Mint
14-07-2025
- Health
- Mint
US Man uses ChatGPT to lose 11 kilos in 46 days, shares fitness routine and diet
A 56-year-old man from the United States is going viral for shedding 11 kilos in just 46 days, with no trainer, no gym membership, and no fad diets. His secret? ChatGPT. Cody Crone, a YouTuber based in the Pacific Northwest, turned to AI for guidance after struggling with weight gain and physical discomfort. On his 56th birthday, feeling unhappy with his health and fitness, Cody decided it was time to make a change and let AI build his personalised weight loss plan. From 95 kg to 83 kg, Cody lost 25.2 pounds (approximately 11.4 kg) in just a month and a half. What makes his story stand out is that he didn't use weight loss drugs like Ozempic or hire a personal coach. Instead, he relied on discipline, clean eating, smart supplementation, and consistent workouts- all planned with help from AI. Cody shared his transformation and routine in a YouTube video. Here's what his AI-guided plan looked like: Two whole-food meals a day, with a long fasting window. No food after 5 PM. Meals were completely free from processed food, sugar, seed oils, and dairy. Breakfast: 4 eggs, half a pound of lean grass-fed beef, steel-cut oats (unsweetened), and a greens supplement. Dinner: 1/3 cup jasmine rice, 8 oz (approx 225g) of lean steak, olive oil or half an avocado. No snacks or sugar-laden treats. Included creatine, beta-alanine, whey protein, collagen, magnesium, and other clean-label performance boosters. Built a home garage gym with a pull-up bar, resistance bands, kettlebells, dip bar, and a weighted vest. Workout at 6 AM daily, lasting 60–90 minutes, six days a week. Woke up at 4:30 AM every day to maintain consistency. Prioritised 7–8 hours of sleep. No screens an hour before bed, blackout curtains, and even raw honey before bedtime to aid sleep quality. No synthetic bed linen or electronics in the bedroom. Drank around 4 litres of water daily, stopping by early evening. Got 15–20 minutes of morning sunlight every day to support metabolism and energy. Recorded his fasted weight every morning, allowing AI to tweak the plan based on real-time progress. Cody said he not only lost weight but also gained strength, reduced joint pain, improved his sleep, and experienced clearer thinking and better mental health.
&w=3840&q=100)
Business Standard
31-05-2025
- Business
- Business Standard
Best of BS Opinion: How the calm today may be masking deeper risks
You know that moment when your car hums along just fine, but one weird little light on the dashboard keeps flickering? That quiet, pulsing reminder that even though everything feels normal, something might still be off under the hood? That's how the world also feels right now, technologies are advancing, markets are steadying, systems are stabilising, but those little blinking lights? They're everywhere. Let's dive in. Take the bond market. Nir Kaissar walks us through a reality-check: people are panicking over interest rates, but zoom out and the current 4.5 per cent on 10-year Treasuries is historically normal. The bigger issue? We're just not used to it anymore after years of easy money. The machine is working but fiscal red flags like $5 trillion in potential deficits or unsustainable tax cuts are blinking. Ignore them, and we might end up stalling on the highway. Meanwhile, Mihir S Sharma cautions that with artificial intelligence, the system isn't just running, it's accelerating like a Tesla in Ludicrous Mode. But no one agrees on where it's heading. Some say AI will democratise creativity and generate jobs. Others predict mass layoffs and existential threats. Will it empower developing countries or further divide them from AI-rich superpowers? The engine is sleek, but no one can read the road signs ahead. And for Pakistan, Shekhar Gupta points out a particularly worrisome light on the regional dashboard: Field Marshal Munir. A military chief with unprecedented power, a hollowed-out civilian government, and a jailed popular rival (Imran Khan) — this isn't a hybrid regime anymore, it's a duckbilled platypus of power. Past performance, as they say in mutual fund ads, is no guarantee of future returns. But in Pakistan, past power-hungry generals have ended up disgraced, exiled, or dead. Munir, however, seems poised to act while he still can. India, beware: the system next door may look stable, but it's humming toward a potentially explosive 12 months. Then there's Devangshu Datta, who shows how GPS and drones, marvels of civilian tech, are now shaping military arsenals. From V-1 flying bombs to AI-guided kamikaze drones, our tools for convenience are doubling as tools for conflict. Precision strikes and geolocation warfare are here, whether we're ready or not. The system works brilliantly, for both Swiggy and the battlefield. Even climate scientists are leaning into geoengineering, as Kumar Abishek writes. Solar Radiation Modification might temporarily cool the Earth by bouncing sunlight back into space. It's technically feasible, increasingly funded, and yet deeply risky. The cooling may come with unintended consequences and no global playbook to manage them. The science is on, but the ethics light is blinking furiously.
