Predicting Flight Delays and Deepfake Education

Aug 9

Emerging Tech Roundup — August 9

The Quantious team’s top picks for timely trending news in the tech world.

This week in tech: Microsoft shares quiz to educate people on deepfake AI images, Nat Geo shares first Vision Pro immersive environment, Flighty’s popular flight-tracking app can now predict delays, Amazon Music’s new AI feature lets users browse podcast episodes by topic, Nest Learning Thermostat had a major upgrade, OpenAI released top feature requested by developers, and new wearable tech works even when bent or twisted.

Real or Not? New Microsoft Quiz Targets Deepfake AI Images and We Still Can’t Really Tell the Difference

(Geek Wire, August 5)

Microsoft President Brad Smith shared a picture of a dog on LinkedIn to highlight a new Microsoft-generated quiz that challenges users to distinguish between real and AI-generated images. This “Real or Not” quiz aims to improve AI literacy, as deepfakes — AI-created images, videos, or audio used to spread false information — become increasingly difficult to detect. Microsoft also provides educational resources with tips for identifying and reporting deepfakes to help ensure election integrity and media accuracy.

Nat Geo’s First Vision Pro Immersive Environment Takes You to Iceland

(Engadget, August 5)

National Geographic is embracing new media by delivering its first immersive experience for the Apple Vision Pro, featuring Iceland’s Thingvellir National Park on Disney+. This allows viewers to enjoy 3D versions of The Avengers, The Age of Ultron, and the first two Ant-Man movies amidst stunning natural landscapes. While a dedicated big-budget National Geographic experience for the Vision Pro isn’t expected soon, adapting existing VR projects or unique helmet projections remains a possibility.

Flighty’s Popular Flight-Tracking App Can Now Predict Delays Using Machine Learning

(Tech Crunch, August 6)

Flighty has launched Flight 4.0, using machine learning and aviation authority data to provide early warnings and specific reasons for flight delays. The app tracks major delay causes like late aircraft and airspace issues, informing users ahead of airlines to help them make better travel decisions. By monitoring various delay factors, Flighty offers clear notifications such as “Delayed 45m. Due to late arriving aircraft from New York.”

Amazon Music’s New AI Feature Lets You Browse Podcast Episodes by Topic

(Tech Crunch, August 6)

Amazon Music has launched a new AI-powered feature called “Topics,” enabling users to explore podcast episodes by specific discussed topics. The AI analyzes podcast transcripts to identify key topics, which are then tagged by human team members, allowing users to find related episodes on themes like “Coffee” and “Dopamine.” Available now for U.S. users on iOS and Android, the feature will initially cover top podcasts and plans to expand to more content over time.

The Nest Learning Thermostat Gets its Biggest Upgrade in Over a Decade

(The Verge, August 6)

Google has announced the Nest Learning Thermostat (4th gen) with a new design, enhanced features, and support for the smart home standard Matter, enabling native compatibility with Apple Home. Priced at $279.99, it is more expensive than its predecessor but includes a new Nest Temperature Sensor (2nd gen) for monitoring multiple rooms. The updated model uses AI for better learning of heating and cooling schedules, incorporates a Soli radar sensor for presence detection, and supports Matter over Wi-Fi — though it lacks a Thread radio.

OpenAI Has Finally Released the №1 Feature Developers Have Been Desperate For

(Venture Beat, August 6)

OpenAI has released the highly anticipated Structured Outputs in its API, addressing a major developer concern. This feature ensures that model-generated outputs match JSON Schemas, improving consistency and reliability in applications. OpenAI CEO Sam Altman confirmed the release was driven by popular demand, and evaluations with the new GPT-4o show a perfect 100% accuracy.

New Wearable Tech Works Even When Bent, Twisted, and Stretched Up to 150%

(Interesting Engineering, August 7)

Researchers have developed a new approach to enhance wearable performance. This innovation — which layers rigid integrated circuits onto a flexible substrate — addresses the limitations of current stretchable devices. By creating heterogeneous rigidity-protective structures, the team successfully balanced flexibility and stability — ultimately enabling advanced motion capture wearables with built-in machine learning.