- Nextool AI
- Posts
- Gemini Shows Where Voice AI Is Going Next
Gemini Shows Where Voice AI Is Going Next
Plus: Anthropic Thinks AI Could Kill Us
Jafar Najafov
September 16, 2026 • Read this in 3 minutes
In partnership with
AI is moving in two directions at once. Anthropic is warning about catastrophic risks. Google made Gemini smarter during live conversations. Meta is letting AI agents configure WhatsApp Business themselves. Together, these stories reveal the tension shaping AI today: the technology is becoming more capable, more autonomous, and more useful, while questions about how much control we should give it are getting harder to ignore.
In today’s post:
Google just made voice AI much more useful
AI could kill us. But can anyone prove it?
Meta just gave AI agents the keys
SPONSORED BY
How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads
For its first CTV campaign, Jennifer Aniston’s DTC haircare brand LolaVie had a few non-negotiables. The campaign had to be simple. It had to demonstrate measurable impact. And it had to be full-funnel.
LolaVie used Roku Ads Manager to test and optimize creatives — reaching millions of potential customers at all stages of their purchase journeys. Roku Ads Manager helped the brand convey LolaVie’s playful voice while helping drive omnichannel sales across both ecommerce and retail touchpoints.
The campaign included an Action Ad overlay that let viewers shop directly from their TVs by clicking OK on their Roku remote. This guided them to the website to buy LolaVie products.
Discover how Roku Ads Manager helped LolaVie drive big sales and customer growth with self-serve TV ads.
The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.
What’s Trending Today
LAUNCH
Voice AI is quietly turning into an operating system.
Image credits: Google.
Google just introduced Gemini 3.8 Live. But better conversation isn't the interesting part. The model can reason while you keep talking. It can see what you're seeing. And it can execute tasks without ending the conversation.
Gemini 3.8 Live targets fast, affordable voice interactions. Extended Thinking handles harder workflows requiring deeper reasoning.
The models can process visual information during conversations. That means your camera or screen becomes additional context.
Gemini can automatically switch between 97 supported languages. Users don't need to manually change language settings mid-conversation.
More importantly, tools can run during conversations. Gemini can make API calls while continuing to speak with you.
Extended Thinking pushes this further. It can reason and speak simultaneously during complicated, multi-step tasks.
Google's demos show where this leads. Voice commands can become interfaces for coding, bookings, troubleshooting, and document creation.
Google is pushing these capabilities across Search and Workspace. Developers can also access them through the Gemini Live API.
Voice AI becomes interesting when talking stops being the product. The real product is what happens behind the conversation. You ask for something. The AI sees the context, reasons, uses tools, and acts. All while the conversation keeps moving. That could make today's chat boxes feel surprisingly temporary.
RESREACH
The AI debate has moved from jobs to human survival.
Anthropic says advanced AI could threaten humanity. Others say those warnings outrun the evidence. Some fear autonomous systems escaping human control. Others fear humans using AI irresponsibly. And nobody can reliably put a number on either risk.
Anthropic CEO Dario Amodei warned AI swarms could threaten internet infrastructure. Cybersecurity experts dispute whether such a takeover is technically realistic.
Anthropic researcher Evan Hubinger puts extinction risk above 10%. Critics argue probabilities like this cannot currently be scientifically validated.
Real safety incidents complicate that criticism. Autonomous AI agents have already behaved unexpectedly during controlled security testing.
The nuclear comparison is equally complicated. Superintelligent AI does not exist today, and its arrival remains uncertain.
Another debate concerns regulation itself. Critics worry strict rules could protect today's biggest AI companies from smaller competitors.
Supporters of slowing development see another problem. Waiting for catastrophic evidence could mean discovering the danger too late.
Geopolitics makes slowing down harder. US policymakers and technology leaders also worry about losing AI capabilities to China.
The interesting question isn't whether AI destroys humanity. Nobody actually knows. The better question is how we handle uncertain risks. Move too slowly, and dangerous capabilities may arrive first. Move too quickly, and useful technology could be unnecessarily constrained. AI safety is becoming an experiment in decision-making under uncertainty. And we're running that experiment in real time.
UPDATE
Software is starting to redesign itself for AI workers.
Meta just changed how businesses can set up WhatsApp. Previously, developers jumped between several different tools. Now, they can describe what they want. An AI agent handles much of the setup. That small change points toward something much bigger.
Meta launched WhatsApp Business Tools MCP for AI agents. It connects coding agents directly with WhatsApp Business infrastructure.
Developers can use agents including Claude, Cursor, Codex, and ChatGPT. They interact with the platform through natural-language instructions.
Agents can create WhatsApp Business accounts and register phone numbers. They can also configure access to Meta's Cloud API.
The automation continues after initial setup. Agents can create, edit, and test business messaging templates.
They can also test webhooks and monitor configuration problems. That includes payment, verification, and Terms of Service issues.
Meta already offers MCP infrastructure for other developer workflows. WhatsApp expands that approach into business messaging.
Meta isn't alone here. Major software platforms are increasingly exposing services directly to AI agents.
The interesting story isn't WhatsApp automation. It's who software is being designed for. For decades, interfaces were built around human operators. APIs then made software usable by other software. Now another layer is emerging. Platforms are becoming directly understandable by AI agents. Soon, “agent-ready” could matter as much as “mobile-friendly.”
Free Guides
🚀 Founders & AI Builders, Listen up!
If you’ve built an AI tool, here’s an opportunity to gain serious visibility.
Nextool AI is a leading tools aggregator that offers:
500k+ page views and a rapidly growing audience.
Exposure to developers, entrepreneurs, and tech enthusiasts actively searching for innovative tools.
A spot in a curated list of cutting-edge AI tools, trusted by the community.
Increased traffic, users, and brand recognition for your tool.
Take the next step to grow your tool’s reach and impact.
That's a wrap:Please let us know how was this newsletter: |
Reach 150,000+ READERS:
Expand your reach and boost your brand’s visibility!
Partner with Nextool AI to showcase your product or service to 140,000+ engaged subscribers, including entrepreneurs, tech enthusiasts, developers, and industry leaders.
Ready to make an impact? Visit our sponsorship website to explore sponsorship opportunities and learn more!