- FryAI
- Posts
- Microsoft takes matters into its own hands
Microsoft takes matters into its own hands
Hunter Kallay
August 31, 2025
In partnership with
Good morning! While the world falls behind the AI trends, we are here to help you stay ahead. 🙌
Swap, Bridge, and Track Tokens Across 14+ Chains
Meet the Uniswap web app — your hub for swapping, bridging, and buying crypto across Ethereum and 14 additional networks.
Access thousands of tokens and move assets between chains, all from a single, easy-to-use interface.
Trusted by millions, Uniswap includes real-time token warnings to help you avoid risky tokens, along with transparent pricing and open-source, audited contracts.
Whether you're exploring new tokens, bridging across networks, or making your first swap, Uniswap keeps onchain trading simple and secure.
Just connect your wallet to get started.
🤯 MYSTERY AI LINK 🤯
(The mystery link can lead to ANYTHING AI-related: tools, memes, articles, videos, and more…)
Today’s Menu
Appetizer: Microsoft launches its first in-house AI models 🦾
Entrée: OpenAI launches “gpt-realtime” for voice AI 🗣️
Dessert: Taco Bell rethinks AI drive-thru after major mishaps 🌮
🔨 AI TOOLS OF THE DAY
🚶♂️ Walk the World: Convert your daily steps into iconic map milestones. → Check it out
👩⚖️ Skala: An intuitive legal platform for startups. → Check it out
MICROSOFT LAUNCHES ITS FIRST IN-HOUSE AI MODELS 🦾
What’s new? Microsoft is rolling out its first in-house AI models.
MAI-Voice-1 is designed to make AI voices sound more natural and engaging, capable of generating a full minute of high-quality audio in under a second. It’s already live in Microsoft’s Copilot Daily and Podcasts, and users can try it hands-on through Copilot Labs for storytelling, meditation, and other creative experiences. | MAI-1-preview is Microsoft’s first-ever foundation model. It’s currently being evaluated on the LMArena platform and will soon roll out across Copilot for text-based tasks, with early testers invited to provide feedback. |
Why is this significant? Until now, Microsoft has largely relied upon OpenAI’s foundation models to power its AI technology, like Copilot. This new transition marks a new era for Microsoft, where they begin separating themselves from reliance on OpenAI. This comes after OpenAI largely failed to satisfy users with its latest GPT-5 release.
OPENAI LAUNCHES “GPT-REALTIME” FOR VOICE AI 🗣️
What’s new? OpenAI has officially launched its Realtime API with major upgrades, including support for phone calling, image inputs, and a new speech-to-speech model called gpt-realtime.
What does this mean? The Realtime API allows developers to build voice agents that sound more natural and respond faster by using a single model to both process and generate speech—rather than stitching together separate systems. The new gpt-realtime model can follow complex instructions, switch languages mid-sentence, and even capture non-verbal cues like laughter. It’s also better at precision tasks, such as repeating alphanumeric codes or reading disclaimers word-for-word. Additionally, two new voices (Cedar and Marin) have joined the lineup, and the system now supports Session Initiation Protocol (SIP), making it possible to connect directly to phone networks.
Fun fact: Think of an “API” like the menu at your favorite fry shop. Instead of guessing how to cook fries from scratch, you just order what you want off the menu. An API works the same way—it’s a set of ready-made “menu options” that lets developers plug powerful AI features (like voice, image, or text tools) straight into their apps. So rather than building the fryer, peeling the potatoes, and salting everything themselves, companies can just “order up” AI capabilities, add their own seasoning, and serve them to their users.
Why does this matter? This upgrade makes AI voice agents far more useful in real-world settings like customer service, healthcare, and real estate. By sounding more human and reacting more intelligently, these agents could simplify tasks that usually require a live representative—saving companies time while making interactions more natural for users.
TACO BELL RETHINKS AI DRIVE-THRU AFTER MAJOR MISHAPS 🌮
What’s new? Taco Bell is reconsidering its use of AI-powered drive-thru ordering after viral clips showed the system making major blunders.
Why? Since 2023, Taco Bell has rolled out voice AI at more than 500 U.S. drive-throughs to cut wait times and reduce mistakes. The technology is designed to take orders directly from customers, but in practice, it has sometimes struggled. Social media is filled with clips of mishaps—like one customer crashing the system by ordering 18,000 water cups, and another frustrated patron whose request for a large Mountain Dew was repeatedly misunderstood. Taco Bell’s Chief Digital and Technology Officer, Dane Mathews, acknowledged that while the AI can impress, it often falters during busy times. He says the company is now rethinking where and when AI should be used, with humans stepping in when needed.
Top Publishers Hand-Selecting Amazon Brands to Promote this Holiday Season
This holiday season, top publishers are handpicking Amazon brands to feature in gift guides, newsletters, and reviews — driving high-intent shoppers straight to storefronts.
Levanta is connecting a select group of 7–9 figure brands with publishers ready to promote products to millions of buyers.
HAS AI REACHED SINGULARITY? CHECK OUT THE FRY METER BELOW:
What do ya think of this latest newsletter? |
Your feedback on these daily polls helps us keep the newsletter fresh—so keep it coming!