- Horizon AI
- Posts
- Real or Fake? Here’s How to Verify If an Image Is AI-Generated 🔍
Real or Fake? Here’s How to Verify If an Image Is AI-Generated 🔍
OpenAI-Backed Chai Discovery Secures $130M to Accelerate AI-Powered Molecule Innovation 🧬
Gina Acosta
December 15, 2025
In partnership with
Welcome to another edition of Horizon AI,
AI image generation is moving fast. Since the release of Nano Banana Pro, highly realistic AI images have been flooding the internet, making it harder than ever to tell what’s real and what’s fake. Today, we show how you can verify if an image is AI-generated in seconds.
Let’s jump into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
Chai Discovery Raises $130M to Accelerate AI Molecule Design
Nvidia Debuts Nemotron 3 Family of Open Models
AI Tutorial: How to check if an image is AI-generated with Gemini
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
CHAI DISCOVERY
Chai Discovery Raises $130M to Accelerate AI Molecule Design
Chai Discovery, an AI-driven drug-discovery startup backed by OpenAI, has raised $130 million to accelerate the development of a “computer-aided design suite” for molecules.
Details:
The Series B round values the company at $1.3 billion and comes just three months after its previous raise, bringing total funding to $225 million.
The round was co-led by Oak HC/FT and General Catalyst, with participation from Thrive Capital, Menlo Ventures, and OpenAI.
Chai Discovery plans to use the funding to build a full “computer-aided design suite” focused on creating and optimizing molecules for drug development.
The raise follows the release of Chai-2, the company’s latest antibody-generation model, which targets diseases that have traditionally been difficult to address with existing approaches.
Co‑founders say 2025 has been “the year of research”, with 2026 expected to focus on real‑world deployment and integration into pharma workflows.
Backers argue AI is badly needed in drug development, where bringing a single medicine to market can take a decade and cost more than $1 billion. This funding positions Chai Discovery to push AI deeper into pharmaceutical pipelines, aiming to shorten timelines, reduce costs, and modernize how new medicines are created.
TOGETHER WITH ROKU
Find customers on Roku this holiday season
Now through the end of the year is prime streaming time on Roku, with viewers spending 3.5 hours each day streaming content and shopping online. Roku Ads Manager simplifies campaign setup, lets you segment audiences, and provides real-time reporting. And, you can test creative variants and run shoppable ads to drive purchases directly on-screen.
Bonus: we’re gifting you $5K in ad credits when you spend your first $5K on Roku Ads Manager. Just sign up and use code GET5K. Terms apply.
NVIDIA
Nvidia Debuts Nemotron 3 Family of Open Models
Nvidia has introduced a new generation of open-source AI models, positioning them as faster, cheaper, and more capable as open models from Chinese labs gain wider adoption.
Details:
Nvidia announced the third generation of its Nemotron large language models, designed for writing, coding, and multi-step reasoning tasks. The smallest model, Nemotron 3 Nano, is available now, with larger versions planned for the first half of 2026.
The company says Nemotron 3 Nano is more efficient than earlier versions, lowering compute costs while improving performance on longer and more complex tasks.
While best known for supplying AI chips to companies building closed models, Nvidia also maintains a growing portfolio of open-source models used in areas like simulations and autonomous systems, with companies such as Palantir integrating them into commercial products.
The release comes as open-source models from Chinese firms like DeepSeek, Moonshot AI, and Alibaba gain traction, even as many U.S. governments restrict the use of Chinese models over security concerns.
Nvidia is emphasizing transparency by openly releasing training data and tooling, allowing enterprises and government users to audit, customize, and validate the models for security and reliability.
With some U.S. tech companies like Meta reportedly moving away from open models, Nvidia’s approach positions it as one of the leading American providers still committed to open-source AI.
AI Tutorial
How to check if an image is AI-generated with Gemini
Google has made it easier for users to check if an image is real using Gemini, now that highly realistic Nano Banana Pro images without visible watermarks have flooded social media. Here’s how:
Go to Gemini and sign in.
Tap Add files (+) and upload the suspected image.
Ask a simple question like:
“Is this image made with AI?” or “Is this AI-generated?”
You can also type “@synthid” after attaching an image.
Gemini looks for a SynthID watermark to see if the image was generated or edited with Google AI tools.
AI Tools to check out
🔥 Kling 2.6: Generate high-quality AI videos with audio, multiple aspect ratios, and advanced controls.
🌐 Lufe.ai: AI-powered translation for webpages, PDFs, and images using Gemini, OpenAI, and Claude engines.
📽 FlovaAI: Create cinematic and social media videos with AI. Text, audio, storyboard, and editing—all together.
📝 Contenov: Generate comprehensive SEO blog briefs faster and smarter using Artificial Intelligence.
👉 Dex: Browser extension that proactively organizes tabs, shortcuts workflows, and handles tasks for you.
TOGETHER WITH ROKU
Shoppers are adding to cart for the holidays
Peak streaming time continues after Black Friday on Roku, with the weekend after Thanksgiving and the weeks leading up to Christmas seeing record hours of viewing. Roku Ads Manager makes it simple to launch last-minute campaigns targeting viewers who are ready to shop during the holidays. Use first-party audience insights, segment by demographics, and advertise next to the premium ad-supported content your customers are streaming this holiday season.
Read the guide to get your CTV campaign live in time for the holiday rush.
AI Findings/Resources
🏫 AI in education amplifies human connection by freeing teachers and personalizing learning.
💻 Tip for vibe coders: always ask for a regression test after a bug fix.
👀 The same prompt was run on GPT-5.2, Gemini 3.0, and Opus 4.5. Which model did it best?
The latest in AI and Tech
The startup has raised $2.5 million in seed funding to grow Momo Self Care, an AI companion app designed to help users build habits. The app features a digital pet that reminds users to complete tasks, rewards progress, and suggests self-care habits based on goals. Backers include a16z speedrun, SignalFire, and True Global.
Disco is a new Gemini-powered experiment that turns open browser tabs into custom web apps called “GenTabs.” The tool suggests and builds interactive apps based on what you’re browsing, using Gemini 3 and your browsing context. Users can refine these apps with natural language and link results back to original sources. It is launching as a limited test via Google Labs, with access through a waitlist for macOS users.
Meta has partnered with ElevenLabs to bring AI-powered audio features to platforms like Instagram and Horizon. The collaboration will support dubbing Reels into local languages, generating music, and creating character voices for Horizon’s immersive experiences.
The new tool allows developers and designers to drag and rearrange interface elements while AI agents automatically update the underlying code. In addition to making manual changes, users can also request edits from Cursor’s AI agent using natural language.
Google has updated its Live with Search feature to use its newest Gemini model for native audio. The change improves voice conversations in AI Mode within the Google app, making responses sound “more fluid and expressive.”
That’s a wrap!
We'd love to hear your thoughts on today's email!Your feedback helps us improve our content |
Not subscribed yet? Sign up here and send it to a colleague or friend!
See you in our next edition!
Gina 👩🏻💻