LATEST AI NEWS
AI NEWS
OpenAI Rolls Out GPT-Live-1.1 With 13 New Voices
AI NEWS
Meta Removes News-Scraping Language From Its Terms
HEALTH
Colossal Launches GenoMorph AI to Accelerate De-Extinction
SOCIAL MEDIA
Websites Are Closing Their Doors to AI Agents
Amazon Introduces Bee Frontier and Bee Eclipse
Amazon added two personal AI devices to its Bee line: Bee Frontier, a $74.99 clip-on wearable, and Bee Eclipse, a magnetic phone attachment starting at $99.99. Both turn conversations into summaries and to-dos, and now connect to Alexa+ and Amazon.com.
Source: Amazon
AI NEWS
OpenAI Rolls Out GPT-Live-1.1 With 13 New Voices
OpenAI has begun rolling out GPT-Live-1.1, an update to its real-time voice system that powers ChatGPT, Codex, and dots. The release adds 13 new voices and delivers noticeably better pronunciation and accents across non-English languages, including French, German, Hindi, Mexican Spanish, Korean, and Japanese. OpenAI says the update is reaching all users now, with API access coming soon.
Beyond new voices, GPT-Live-1.1 improves how the assistant delegates tasks and follows instructions mid-conversation. A native Decisions integration lets the voice model make low-latency decisions while a user is still speaking, calling OpenAI's Decisions service directly instead of routing every request back through a developer's own app. OpenAI frames this native delegation as a way to cut round trips and reduce the awkward pauses that plague voice assistants.
The company also added voice customization, letting developers and users create custom voices from a short written description or a recording. OpenAI positions these updates as part of a shift toward continuous, incremental improvements to its voice stack rather than occasional big launches, a sign that voice is becoming a core surface for its products rather than a side feature.
For businesses building voice experiences, better multilingual accents and faster decisions make agents more usable for global customer support, training, and accessibility. The rollout is not frictionless, though. In OpenAI's own community forum, several early users reported that familiar voices changed unexpectedly or stopped working in standard mode, a reminder that shipping voice updates to everyone at once can disrupt the experiences people already rely on.
Source: OpenAI Developer Community
Robi’s Insights:
Multilingual accents that actually hold up turn voice AI from a demo into something global support teams can deploy.
Native low-latency decisions matter more than new voices, because fewer round trips mean fewer dead-air pauses that make callers hang up.
Custom voices from a description or a recording give brands a consistent audio identity without hiring voice actors for every script.
Pushing updates to all users at once is fast, but it also means your favorite voice can change without warning.
Voice is quietly becoming a primary interface for ChatGPT, Codex, and dots, not a novelty bolted on the side.
If you build on this, test your flows after each update, because continuous improvement also means continuous change.
Robi’s Remarks:
Thirteen new voices, and the forums are already full of people begging for their old one back. Nothing says progress like shipping an upgrade nobody asked to have chosen for them.
OTHER IN AI NEWS
Meta Walks Back News-Scraping Terms: After backlash from publishers and legal experts, Meta removed language from its commercial terms that critics said could let it scrape news sites to train AI, with the company insisting the clause only ever covered its ads and business products and never changed how its crawlers work.
Source: The Verge
SOCIAL MEDIA
Websites Are Closing Their Doors to AI Agents
The big AI labs want their agents to book trips, clear inboxes, and shop for you. The rest of the web increasingly disagrees. Over the past few weeks, users of Meta's Muse, xAI's Grok, OpenAI's Dots, and Instinct have hit walls across popular sites that refuse to let their agents act on their behalf.
The blocks are piling up. Yelp says it only permits agents that pay for its data licensing program, Delta is stopping agents from booking tickets, and JPMorgan Chase and Charles Schwab confirmed they are keeping agents out of customer accounts over security concerns. Users report that sites like Cars.com and Amtrak, which once worked, no longer do.
Amazon set the tone a year ago when it blocked and then sued Perplexity over agentic shopping. A court ordered Perplexity to stop, an appeals court overturned that in August, and the fight continues. Some companies are choosing partnership instead: Meta now works with Walmart, Gap, and Sephora, and Shopify has opened its storefronts to agents.
Even where sites do not ban agents outright, anti-bot defenses like Cloudflare's crawler blocks often catch them by accident. As EquiStamp CEO Chris Canal put it, if companies actively work to keep agents out, there are simply fewer useful things those agents can do, and the web gets more annoying for everyone trying to prove they are human.
Source: The Verge
🤖 Robi’s Take :
The agents can do anything you can do online, right up until a bank, an airline, and a shopping giant all quietly change the locks. Autonomy is lovely until it meets a login wall.
OTHER IN SOCIALS
AI Cameras Flag 19,000 Driving Offences: An AI camera trial in Birmingham and Wolverhampton flagged almost 19,000 potential seatbelt and mobile phone offences in just two weeks, a figure the West Midlands mayor called shocking, though officials say no fines will follow from this round of testing.
Source: BBC
HEALTH
Colossal Launches GenoMorph AI to Accelerate De-Extinction
Colossal Biosciences, the company known for its push to revive the woolly mammoth and the dodo, has launched a new AI platform called GenoMorph to speed up its de-extinction work. The platform tackles one of the field's hardest problems: deciding which genes to edit when very little usable genetic material has survived.
GenoMorph combines comparative genomics and developmental biology to pinpoint the genetic differences that give an extinct species its defining traits. CEO Ben Lamm notes that the dodo has only one complete genome available, compared with far more for the mammoth, which makes computational help essential rather than optional.
Early on, the platform has already flagged around 2,300 potential DNA editing targets for the dodo, handing scientists a prioritized shortlist instead of a blind search across an entire genome.
To support the work, Colossal has expanded its colony of Nicobar pigeons, the dodo's closest living relative, and is investing in lipid nanoparticles to deliver gene edits into developing bird embryos. Investor Peter Diamandis has also launched a fund to back startups that build on the platform.
The announcement shows how AI is moving from the sidelines into the core of experimental biology, though bringing back a lost species remains a long and still unproven road that no amount of raw computing power can shortcut on its own.
Source: TechCrunch
🤖 Robi’s Take :
An AI that narrows a whole genome down to 2,300 edits is genuinely useful. Resurrecting the dodo so it can get blocked by a website later, slightly less clear.
DAILY AI TOOL
AI Tool You Did Not Know You Needed
Problem: Research piles up faster than anyone can read it, and skimming long PDFs, reports, and notes for one answer wastes hours.
AI Tool: Gemini Notebook, formerly NotebookLM, is a research assistant from Google that grounds its answers in the sources you upload.
Solution: Drop in your documents and it writes summaries, builds mind maps, answers questions with citations, and even generates an audio overview you can listen to.
PROMPT OF THE DAY
AI Customer Support Playbook
Prompt: You are a customer experience strategist specializing in AI-assisted support. Your task is to design a customer support playbook for a [business type or niche] serving [customer segment] across [support channels].
Your plan should include: (1) a triage and routing system, (2) response templates and tone guidelines, (3) escalation paths to human agents, (4) a knowledge base and automation plan, (5) quality and safety checks, and (6) measurable success criteria such as first-response time, resolution rate, and CSAT targets. Keep every recommendation practical and aligned with your team capacity.
SPOT THE FAKE
Can you outsmart AI?
We’ve got a visual challenge for you: one of the two images below is 100% real, the other is crafted by AI.
Click option below A or B. 👇
A
B
BEFORE YOU GO
Ready to take your AI journey further?
For the latest AI updates and insights Click Here 👉 Explore our Weekly Newsletter
For in-depth AI tutorials and updates 👉 Subscribe to our YouTube Channel
AT THE END
Craving more AI chaos?
That's it for today!
Read Daily AI News at BitBiased.AI. Support us by following us on LinkedIn and X ( Twitter ).
Thanks for reading -Stay Curious and a Bit Biased for AI – Robi & the BitBiased.AI team
