AI News Sep 16, 2026
By Frontier Editorial •
Key Takeaways
- Google launches Gemini 3.8 Live audio models starting at $1.38 per hour: Google DeepMind has released Gemini 3.8 Live and 3.8 Live Extended Thinking, two new audio models for devel…
- Nvidia and Palantir reportedly scale back Anthropic model use over data logs: Despite assurances from OpenAI and Anthropic that customer data is not used for training, AI companies…
- Salesforce and Nvidia launch Koa reasoning model for business tasks: Salesforce has introduced Koa, a new reasoning model developed in collaboration with Nvidia. Built on Nvidia's …
- OpenAI: $1.2 trillion (OpenAI)
- 1. Copy the structured prompt template from the source text.
What are the top AI breakthroughs?
This Sep 16, 2026 covers 9 curated AI news items spanning technology, research, and product developments. Google launches Gemini 3.8 Live audio models starting at $1.38 per hour: Google DeepMind has released Gemini 3.8 Live an...
Google launches Gemini 3.8 Live audio models starting at $1.38 per hour: Google DeepMind has release…
Google launches Gemini 3.8 Live audio models starting at $1.38 per hour: Google DeepMind has released Gemini 3.8 Live and 3.8 Live Extended Thinking, two new audio models for developers that lead the Artificial Analysis speech-to-speech leaderboard. Priced at $1.38 per hour of voice conversation, the models significantly undercut OpenAI's GPT-Live-1.
Nvidia and Palantir reportedly scale back Anthropic model use over data logs: Despite assurances fro…
Nvidia and Palantir reportedly scale back Anthropic model use over data logs: Despite assurances from OpenAI and Anthropic that customer data is not used for training, AI companies continue to face data trust issues. Palantir, Nvidia, and Booz Allen Hamilton reportedly scaled back their use of Anthropic's Fable model after learning it stores usage logs for 30 days.
Meta reportedly plans to launch camera-free smart glasses this fall: Meta Platforms is planning to l…
Meta reportedly plans to launch camera-free smart glasses this fall: Meta Platforms is planning to launch a camera-free model of its smart glasses this fall, according to sources familiar with the matter. The upcoming device will reportedly feature six microphones to enable users to interact with Meta's AI chatbot and Muse.
Agility Robotics unveils Digit 5 humanoid robot to work alongside humans: Agility Robotics has intro…
Agility Robotics unveils Digit 5 humanoid robot to work alongside humans: Agility Robotics has introduced Digit 5, the latest version of its humanoid robot designed for warehouses and factories. The company states that the new model is capable of working directly alongside human employees without the need for safety fences.
Meta launches Meta One subscription bundles globally with extra AI usage: Meta has launched its new …
Meta launches Meta One subscription bundles globally with extra AI usage: Meta has launched its new Meta One subscription bundles globally, pairing standalone app subscriptions with extra AI usage following the release of its Muse AI assistant. Available in several tiers for individual users, creators, and businesses, some of these bundles were previously in testing earlier this year.
Salesforce and Nvidia launch Koa reasoning model for business tasks: Salesforce has introduced Koa, …
Salesforce and Nvidia launch Koa reasoning model for business tasks: Salesforce has introduced Koa, a new reasoning model developed in collaboration with Nvidia. Built on Nvidia's open-weight Nemotron model, Koa is specifically trained to perform sales, marketing, and customer-support tasks.
Meta releases WhatsApp Business MCP server for AI-managed setup: Meta has released a new WhatsApp Bu…
Meta releases WhatsApp Business MCP server for AI-managed setup: Meta has released a new WhatsApp Business Model Context Protocol (MCP) server. This tool allows developers to use AI coding agents such as Claude, Cursor, Codex, and ChatGPT to manage setup, messaging templates, testing, and troubleshooting.
Developers bypass Anthropic to run GPT-5.6 Sol via Claude Code: Some developers are using the Claude…
Developers bypass Anthropic to run GPT-5.6 Sol via Claude Code: Some developers are using the Claude Code harness to access cheaper, non-Anthropic models via proxies and services like OpenRouter. This includes using the tool to run OpenAI's GPT-5.6 Sol. The workaround was highlighted by Boris Cherney, the head of Anthropic's coding tool.
Mozilla report shows paid AI models offer only a four-month lead over open source: A preview of Mozi…
Mozilla report shows paid AI models offer only a four-month lead over open source: A preview of Mozilla's upcoming report reveals that paying for proprietary frontier AI models only secures a four-month head start in capabilities compared to open-source alternatives. The report highlights how rapidly cheap open models are catching up to their expensive proprietary counterparts.
What are the latest AI investment signals?
Latest AI investment signals: 5 funding rounds, 0 market updates, and 0 M&A transactions.
Primary Market – Funding Rounds
| Company | Amount | Round |
|---|---|---|
| OpenAI | $1.2 trillion | OpenAI |
| Gates Foundation | 1000000000 | Bill Gates |
| TypeSafe AI | 40000000 | seed |
| Artificial Intelligence Underwriting Company | $40 million | Series A |
| Exein | $270 million | Exein |
Secondary Market – Market Updates
No secondary market data.
M&A – Mergers & Acquisitions
No M&A data.
What are practical AI tips this week?
2 practical AI tips curated from Reddit communities and expert blogs. Prompt compresses long reports into structured one-page briefs...
Prompt Engineering
Prompt compresses long reports into structured one-page briefs
1. Copy the structured prompt template from the source text. 2. Paste your long report into the designated report section of the prompt. 3. Run the prompt in your preferred AI model to generate a structured, one-page brief.
Prompt Engineering
Reusable prompt system turns bullet notes into natural prose
1. Set the provided writing rules as a system prompt in your AI tool. 2. Input your rough, fragmented bullet notes into the chat. 3. Generate clean, natural paragraphs that preserve your exact meaning without AI buzzwords.