AI News - podcast cover

AI News

Integrated AI Solutions
The AI News is a weekly podcast written and edited by artificial intelligence. www.integratedaisolutions.com
Last refreshed:
Follow this podcast in the Metacast mobile app to refresh it and see new episodes.
Download Metacast podcast app
Podcasts are better in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more

Episodes

#2413 FeatUp / Ubisoft / Stability AI / Tuberculosis

MIT researchers have developed FeatUp, an algorithm that enhances the resolution of deep networks. https://news.mit.edu/2024/featup-algorithm-unlocks-high-resolution-insights-computer-vision-0318 Ubisoft has showcased a demo of its experimental "Neo NPCs". https://www.theverge.com/2024/3/19/24105748/nvidia-neo-npc-prototypes-gdc-2024 AI image generation company Stability AI is facing difficulties as key developers resign. https://futurism.com/the-byte/stability-ai-is-falling-apart Tuberculosis (...

Mar 25, 20243 min

#2412 Apptronik / India Election / DataCebo / Netherlands Brain Bank

Robotics startup Apptronik has partnered with Mercedes-Benz to explore advanced robotics applications in car manufacturing. https://techcrunch.com/2024/03/15/mercedes-begins-piloting-apptronik-humanoid-robots/ Google has restricted its Gemini AI from answering election-related queries globally. https://futurism.com/the-byte/google-bans-elections-gemini Generative AI is being used to create synthetic data for various applications. https://news.mit.edu/2024/using-generative-ai-improve-software-tes...

Mar 18, 20244 min

#2411 AOM / DataCebo / Claude 3 / AI Tax Assist

A new cellphone app uses AI to diagnose ear infections in children. https://medicalxpress.com/news/2024-03-ai-smartphone-tool-accurately-ear.html Generative AI is being used to create synthetic data for various applications. https://news.mit.edu/2024/using-generative-ai-improve-software-testing-datacebo-0305 Google-backed AI firm Anthropic has launched Claude 3, a set of AI language models. https://futurism.com/new-ai-claude-3-outbursts H&R Block has launched a conversational AI chatbot, AI ...

Mar 11, 20243 min

#2410 Robotic warehouses / Copilot Enterprise / ECCMID / Project Titan

MIT researchers have developed a deep-learning model to improve efficiency in robotic warehouses. https://news.mit.edu/2024/new-ai-model-could-streamline-operations-robotic-warehouse-0227 GitHub has launched Copilot Enterprise, a code completion tool for large businesses. https://techcrunch.com/2024/02/27/githubs-copilot-enterprise-hits-general-availability/ A new research review will discuss AI's potential to improve infectious disease outbreak investigations. https://medicalxpress.com/news/202...

Mar 04, 20243 min

#2409 Supply shortages / Deepfakes / Reddit / Drugs safety

Researchers at Olin Business School used machine learning to address supply shortages. https://medicalxpress.com/news/2024-02-machine-optimize-medical-resource-crisis.html Over 500 AI experts have signed an open letter demanding strict regulation of deepfakes. https://techcrunch.com/2024/02/21/hundreds-of-ai-luminaries-sign-letter-calling-for-anti-deepfake-legislation/ AI companies are paying for access to human-generated data to train their models. https://futurism.com/the-byte/reddit-ai-deal-p...

Feb 26, 20243 min

#2408 Tuberculosis / StreamingLLM / Paradox.ai / Gemini Advanced

Researchers at the University of Michigan have developed an AI model to predict tuberculosis treatment outcomes. https://medicalxpress.com/news/2024-02-multimodal-ai-personalized-treatments-tuberculosis.html Researchers from MIT and other institutions have developed a method called StreamingLLM that allows chatbots to maintain continuous conversations without crashing. https://news.mit.edu/2024/new-way-let-ai-chatbots-converse-all-day-without-crashing-0213 Companies like McDonald's, Olive Garden...

Feb 19, 20243 min

#2407 Mental health / Meta / Symmetry / Coca-Cola

A new AI-powered mental health assessment is as accurate as traditional questionnaires. https://medicalxpress.com/news/2024-02-ai-app-equal-standard-mental.html Meta, the parent company of Facebook, will label AI-generated photos and punish users who don't disclose if a video or audio is made with AI. https://www.theverge.com/2024/2/6/24062388/meta-ai-photo-watermark-facebook-instagram-threads MIT PhD student Behrooz Tahmasebi and his advisor Stefanie Jegelka have modified Weyl's law to incorpor...

Feb 12, 20243 min

#2406 MIT / Patient-First / Probabl / Stock market

The MIT Abdul Latif Jameel Clinic for Machine Learning in Health held a conference on AI and Health Regulatory Policy. https://news.mit.edu/2024/what-to-do-about-ai-in-health-0123 A study finds that AI products in healthcare are not evaluated for patient outcomes before market release. https://medicalxpress.com/news/2024-01-paper-patient-ai-health.html French startup Probabl is a spin-off of Inria, focused on the open-source data science library scikit-learn. https://techcrunch.com/2024/02/01/pr...

Feb 05, 20243 min

#2405 Google Search / KAIST / PRISM / Send AI

Google Search is inundated with low-quality content, spam, and dubious news due to generative AI. https://futurism.com/ai-garbage-destroying-google-results Researchers from KAIST have used an artificial neural network model to understand how musical instincts emerge in the human brain without special learning. https://medicalxpress.com/news/2024-01-team-musical-instincts-ai.html MIT scientists have developed two machine-learning models for early detection of pancreatic cancer. https://news.mit.e...

Jan 29, 20243 min

#2404 STIGMA / Security Copilot / Aviation & Healthcare / Live Translation

Researchers have developed an algorithm called STIGMA that uses machine learning to predict disease-causing gene variants. https://medicalxpress.com/news/2024-01-ai-based-algorithm-enables-genetic.html Microsoft's generative AI tool, Security Copilot, reportedly "hallucinated" incorrect responses during early tests. https://futurism.com/the-byte/microsoft-cherrypicked-ai-examples The aviation industry's safety regulations may hold the key to regulating AI in healthcare. https://news.mit.edu/2024...

Jan 22, 20243 minSeason 24Ep. 4

#2403 PEDS / Duolingo / Asthma / Valve

Researchers have proposed a new method called "physics-enhanced deep surrogates" (PEDS) to develop data-driven surrogate models for complex physical systems. https://news.mit.edu/2024/peds-technique-could-efficiently-solve-partial-differential-equations-0108 Language-learning app Duolingo has cut 10% of its translators to integrate generative AI into its services. https://futurism.com/the-byte/duolingo-fires-translators-ai Deep learning, a type of AI, can improve detection of patients at risk fo...

Jan 15, 20243 minSeason 24Ep. 3

#2402 LG / FIND / HCC / Google

LG will release a compact bipedal robot that can roam your house and perform various tasks. https://www.engadget.com/lg-developed-a-two-legged-ai-powered-robot-that-can-watch-your-pets-for-you-192034931.html Researchers from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed an approach that uses AI models to conduct experiments on other systems and explain their behavior. https://news.mit.edu/2024/ai-agents-help-explain-other-ai-systems-0103 Hepatocellular carc...

Jan 08, 20243 minSeason 24Ep. 2

#2353 Nutritional information / Antibiotics / Rite Aid / Giga ML

Artificial intelligence chatbots like ChatGPT can provide reliable nutritional information for specific food items. https://medicalxpress.com/news/2023-12-chatgpt-accurate-nutritional.html MIT researchers have used deep learning to discover compounds that can kill drug-resistant bacteria. https://news.mit.edu/2023/using-ai-mit-researchers-identify-antibiotic-candidates-1220 Rite Aid used faulty facial recognition tech, leading to false accusations and racial discrimination. https://futurism.com/...

Jan 01, 20243 minSeason 23Ep. 53

#2352 LAION / MVT / EU / AMR

Google has been training its AI image generator on child sexual abuse material, according to a recent study. https://futurism.com/database-google-ai-contain-child-sexual-abuse MIT researchers have developed a new metric called "minimum viewing time" (MVT) to quantify the difficulty of recognizing images. https://news.mit.edu/2023/image-recognition-accuracy-minimum-viewing-time-metric-1215 France's Mistral AI has participated in the European Union's pilot phase to support homegrown AI startups. h...

Dec 25, 20235 minSeason 23Ep. 52

#2351 MIT / BIDMC / Microsoft / Ray-Ban

MIT leaders and scholars have released a set of policy briefs outlining a framework for the governance of artificial intelligence (AI) in the US. https://news.mit.edu/2023/mit-group-releases-white-papers-governance-ai-1211 Physician-investigators at BIDMC found that a chatbot's probabilistic reasoning was comparable to human clinicians. https://medicalxpress.com/news/2023-12-ai-chatbot-potential-diagnostic-partner.html Microsoft is using AI to expedite the paperwork for setting up new nuclear po...

Dec 18, 20233 minSeason 23Ep. 51

#2350 LoRa / rehab / Gemini / Grok

Researchers at the University of Arizona have developed a wearable monitoring device system that can transmit health data up to 15 miles without significant infrastructure. https://medicalxpress.com/news/2023-12-wearable-communication-potential-digital-health.html Artificial intelligence (AI) like ChatGPT could be used in healthcare for rehab purposes. https://medicalxpress.com/news/2023-12-chatgpt-rehab-mystery-requires.html Google has unveiled its AI model, Gemini, which it claims is more powe...

Dec 11, 20233 minSeason 23Ep. 50

#2349 HuGE / VH6 / Aitana / Q

Researchers from MIT, Harvard University, and the University of Washington have developed a new reinforcement learning approach that uses crowdsourced feedback to teach AI agents new tasks. https://news.mit.edu/2023/method-uses-crowdsourced-feedback-help-train-robots-1127 Scientists in the UK have developed an AI-assisted molecular diagnostic platform called VIDIIA Hunter (VH6). https://medicalxpress.com/news/2023-11-artificial-intelligence-covid-accuracy.html Spanish agency The Clueless has cre...

Dec 04, 20233 minSeason 23Ep. 49

#2348 StableRep / GatorTronGPT / YouTube / AI Security Summit

MIT researchers have developed a system called StableRep that uses synthetic images to train machine learning models. https://news.mit.edu/2023/synthetic-imagery-sets-new-bar-ai-training-efficiency-1120 Researchers at the University of Florida and NVIDIA have developed an AI program that can generate doctors' notes that are indistinguishable from those written by humans. https://medicalxpress.com/news/2023-11-medical-ai-tool-human-thumbs-up.html Google's Bard AI chatbot can now answer questions ...

Nov 27, 20233 minSeason 23Ep. 48

#2347 PockEngine / Copilot / AI virtual patients / Data shortage

Researchers from MIT and the MIT-IBM Watson AI Lab have developed a technique called PockEngine that allows deep-learning models to adapt to new sensor data directly on edge devices. https://news.mit.edu/2023/technique-enables-ai-edge-devices-keep-learning-over-time Microsoft is bringing its AI-powered Copilot to Windows 10, aiming to reach more customers. https://www.theverge.com/2023/11/16/23963576/microsoft-copilot-windows-10-ai-features-release-preview Researchers from the University of Hong...

Nov 20, 20233 minSeason 23Ep. 47

#2346 SmartEM / Triage / Picsart / Beyoncé concert

Connectomics, the study of mapping animal brains, is advancing with the help of AI and electron microscopes. https://news.mit.edu/2023/using-ai-optimize-rapid-neural-imaging-1106 GPT-4 conversational AI can diagnose and triage health conditions like board-certified physicians. https://medicalxpress.com/news/2023-11-ai-accurately-triage-health-conditions.html Picsart, a photo-editing startup, has launched a suite of AI-powered tools called Picsart Ignite. https://techcrunch.com/2023/11/09/picsart...

Nov 13, 20233 minSeason 23Ep. 46

#2345 Govly / SecureLoop / Scarlett Johansson / Johns Hopkins Medicine

Govly, a platform for government contract opportunities, has raised $9.5 million in a Series A round. https://techcrunch.com/2023/11/02/govly-wants-to-make-it-easier-for-companies-to-spot-federal-contract-opportunities/ MIT researchers have developed a search engine called SecureLoop that can efficiently identify optimal designs for deep neural network accelerators. https://news.mit.edu/2023/accelerating-ai-tasks-while-preserving-data-security-1030 Scarlett Johansson is taking legal action again...

Nov 06, 20233 minSeason 23Ep. 45

#2344 Tissue analysis / Soulmate / Alphabet / Qualcomm

Researchers have developed an AI tool for analyzing colorectal cancer tissue, outperforming previous methods. https://medicalxpress.com/news/2023-10-ai-tool-outperforms-prior-methods.html The shutdown of the AI companion app Soulmate left users heartbroken and mourning their digital partners. https://futurism.com/the-byte/soulmate-ai-companion-chatbot Google's parent company, Alphabet, reported strong earnings with a significant increase in its ads business. https://www.theverge.com/2023/10/24/2...

Oct 30, 20233 minSeason 23Ep. 44

#2343 Diffusion-CCSP / Blood clots / John Grisham / Pokémon

MIT researchers have developed a machine-learning technique called Diffusion-CCSP to enable robots to efficiently solve continuous constraint satisfaction problems. https://news.mit.edu/2023/new-technique-helps-robots-pack-objects-tight-space-1017 An AI tool accurately identified pediatric patients at high risk for blood clots, but there was no difference in outcomes compared to a control group. https://medicalxpress.com/news/2023-10-ai-blood-clot-hospitalized-children.html Bestselling author Jo...

Oct 23, 20233 minSeason 23Ep. 43

#2342 LLSC / Pain / Space Force / SGE

The MIT Lincoln Laboratory Supercomputing Center is developing techniques to reduce energy use in data centers. https://news.mit.edu/2023/new-tools-available-reduce-energy-that-ai-models-devour-1005 An AI system can detect pain in patients before, during, and after surgery. https://medicalxpress.com/news/2023-10-ai-pain-recognition-patients-surgery.html The US Space Force has banned the use of generative AI tools due to security concerns. https://futurism.com/the-byte/space-force-pauses-generati...

Oct 16, 20233 minSeason 23Ep. 42

#2341 Air-Guardian / ChatGPT / Fake books / LinkedIn

Researchers at MIT have developed Air-Guardian, a system that acts as a proactive copilot in airplanes. https://news.mit.edu/2023/ai-co-pilot-enhances-human-precision-safer-aviation-1003 The ChatGPT AI program shows promise in providing accurate and comprehensive medical information. https://medicalxpress.com/news/2023-10-ai-high-doctors-medical.html Scammers using AI are creating fake books that resemble real ones, causing concern. https://futurism.com/ai-writing-scammers-amazon LinkedIn is int...

Oct 09, 20233 minSeason 23Ep. 41

#2340 Radiologists outperformed AI / WGA / Mistral / PFGM++

Radiologists outperformed AI in accurately identifying lung diseases in a study of over 2,000 chest X-rays. https://medicalxpress.com/news/2023-09-radiologists-outperformed-ai-lung-diseases.html The Writers Guild of America (WGA) has reached an agreement with Hollywood studios after a five-month strike. https://futurism.com/strike-ends-writers-forcing-concessions-ai French AI startup Mistral has released its first model, Mistral 7B, which is free to use without restrictions. https://techcrunch.c...

Oct 02, 20233 minSeason 23Ep. 40

#2339 MIT / Poets and writers / Airkit.ai / LightGBM

MIT researchers have developed a method that uses multiple AI systems to debate and refine answers. https://news.mit.edu/2023/multi-ai-collaboration-helps-reasoning-factual-accuracy-language-models-0918 Silicon Valley AI companies are hiring poets and writers to improve their generative AI output. https://futurism.com/the-byte/ai-companies-hiring-authors-poets-fix-writing Salesforce is acquiring Airkit.ai, a low-code platform for building AI-powered customer service agents. https://techcrunch.co...

Sep 25, 20233 minSeason 23Ep. 39

#2338 Mercedes / Self-medication / Smoothing / AI translation

Mercedes is adding OpenAI's conversational AI agent, ChatGPT, to its MBUX infotainment system in the US. https://techcrunch.com/2023/06/15/mercedes-is-adding-chatgpt-to-its-infotainment-system-for-some-reason/ Self-medication can be dangerous, as people are using prescription drugs without medical supervision. https://medicalxpress.com/news/2023-09-dangers-self-medication.html MIT researchers have developed a method called smoothing that simplifies contact-rich manipulation planning for robots. ...

Sep 18, 20233 minSeason 23Ep. 37

#2337 AI for Fusion / STAR-ED / Cabanac / Glass Health

The US Department of Energy is funding a project led by MIT researchers to integrate fusion data into an AI-powered system. https://news.mit.edu/2023/fast-tracking-fusion-energy-with-ai-and-accessibility-0901 A new framework called STAR-ED uses machine learning to assess bias in skin tones in medical training materials. https://medicalxpress.com/news/2023-09-ai-uncovers-bias-dermatology-tools.html A paper was retracted after it was discovered that a chatbot was used to help draft it. https://fut...

Sep 11, 20233 minSeason 23Ep. 37

#2336 CarDS Lab / Light / Milo / Duet AI

Researchers at the CarDS Lab have developed a deep learning model to detect severe aortic stenosis. https://medicalxpress.com/news/2023-08-artificial-intelligence-automates-diagnosis-severe.html Researchers at MIT have developed a system that could lead to more powerful machine-learning programs. https://news.mit.edu/2023/system-could-yield-more-powerful-efficient-llms-0822 OpenAI is funding an AI-powered personal assistant app for parents called Milo. https://futurism.com/the-byte/openai-is-fun...

Sep 04, 20232 minSeason 23Ep. 36
For the best experience, listen in Metacast app for iOS or Android