MIT's 'telepathic' wearable arrives

PLUS: OpenAI's new coding model, Google's privacy-first AI

Author

Gen AI Art
September 16, 2025 • Estimated Reading Time: 9 minutes

Together with

An MIT spin-out just unveiled a new wearable device that enables silent communication with your devices. It works by reading subtle muscle signals on your jaw as you silently talk to yourself, interpreting intended speech without a single word spoken aloud.

This technology could offer a private and seamless way to interact with AI, avoiding the social awkwardness of public voice commands. If the device delivers, could it represent the first truly new computer interface since the touchscreen?

Today in AI:

  • MIT’s ‘telepathic’ wearable for silent commands

  • OpenAI's new model specialized for coding

  • Google's new privacy-first AI model

PRESENTED BY 1440 MEDIA

Looking for unbiased, fact-based news? Join 1440 today.

Join over 4 million Americans who start their day with 1440 – your daily digest for unbiased, fact-centric news. From politics to sports, we cover it all by analyzing over 100 sources. Our concise, 5-minute read lands in your inbox each morning at no cost. Experience news without the noise; let 1440 help you make up your own mind. Sign up now and invite your friends and family to be part of the informed.

What’s new? MIT spin-out AlterEgo has unveiled a wearable device that interprets your intended speech by reading subtle neuromuscular signals on your jaw. This allows for silent, voice-free communication with computers and other smart devices.

What matters?

  • The device uses “Silent Sense” technology to detect the internal brain signals sent to your speech muscles when you subvocalize, or silently talk to yourself. This tech was first demonstrated at the MIT Media Lab back in 2018.

  • Early versions of the technology achieved 92% accuracy in transcribing unspoken words from a limited vocabulary, according to the original research paper. The company's initial goal is to help restore voice for individuals with speech-affecting conditions like ALS.

  • Crucially, the system does not read your thoughts. It only captures what you consciously intend to say by purposefully engaging your internal speech system, keeping your private thoughts private.

Why it matters?

This technology offers a private and seamless way to interact with AI, overcoming the social awkwardness of public voice commands. If AlterEgo delivers on its promise, it could represent the first truly new computer interface since the mainstream adoption of the touchscreen.

PROMPT STATION

What’s new? OpenAI has released GPT-5-Codex, a new model fine-tuned for programming that already powers tools like the Codex CLI and VS Code extension.

What matters?

  • The model dynamically adjusts its 'thinking' time based on task complexity, allowing it to tackle more intensive jobs.

  • It introduces new capabilities for automated code review on GitHub repositories, streamlining developer workflows.

  • Performance on code refactoring tasks jumps to 51.3% on OpenAI's internal evaluation, a significant leap from the base GPT-5 model's 33.9%.

Why it matters?

This move provides developers with a more specialized AI assistant for complex programming tasks, directly boosting efficiency. It also signals a broader trend of creating tailored, high-performance AI for specific professional fields.

PRESENTED BY AI REPORT

Used by Execs at Google and OpenAI

Join 400,000+ professionals who rely on The AI Report to work smarter with AI.

Delivered daily, it breaks down tools, prompts, and real use cases—so you can implement AI without wasting time.

If they’re reading it, why aren’t you?

What’s new? Google is pushing forward on two fronts, releasing VaultGemma, a new privacy-focused open model, as its consumer-facing Gemini app surges to the top of the U.S. App Store.

What matters?

  • VaultGemma is built from the ground up with differential privacy to prevent the model from memorizing training data, making it safer for sensitive applications in finance and healthcare.

  • On the consumer front, the Gemini app has become the #1 most downloaded free iOS app in the U.S., surpassing ChatGPT in a significant consumer adoption milestone.

  • This two-pronged strategy demonstrates Google's ability to compete for enterprise and developer trust with secure models while simultaneously capturing the mainstream consumer market.

Why it matters?

This dual approach allows Google to build foundational trust with businesses that require data security. At the same time, its mainstream success with Gemini creates a massive user base that strengthens its competitive position against rivals.

Everything else in AI

Researchers published details on China's SpikingBrain 1.0, a 'brain-like' AI system that runs over 100x faster on domestic MetaX chips by mimicking human neuron activity.

OpenAI agreed to a new revenue-sharing deal with Microsoft that will see its payments shrink from ~20% of revenue to just 8% by 2030, a move reportedly worth over $50B to the AI lab.

California passed AI safety bill SB 53, which now awaits the governor's signature to require major AI companies to disclose safety protocols and establish a public cloud resource called "CalCompute".

Penske sued Google over its AI Overviews, with the Rolling Stone publisher arguing the feature illegally uses its content and harms web traffic.

Essential AI Guides - Reading List:

Let us know!

Work with us

Reach 100k+ engaged Tech Professionals, Engineers, Managers and decision makers. Join brands like MorningBrew, HubSpot, Prezi, Nike, Ahref, Roku, 1440, Superhuman, and others in showcasing your product to our audience. Get in touch now →