- Horizon AI
- Posts
- Dub Your Videos Into Multiple Languages With AI 🌎
Dub Your Videos Into Multiple Languages With AI 🌎
OpenAI Admits ChatGPT Is "Cringe," Launches New Model to Fix It 👀
Gina Acosta
March 04, 2026
In partnership with
Welcome to another edition of Horizon AI,
Reaching a global audience used to take weeks of translation, voice recording, and editing for every language. In today's issue, we'll show you how AI can handle all of that in just minutes.
Let’s jump into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
OpenAI Introduces New ChatGPT 5.3 Instant With Less "Cringe" and Fewer Hallucinations
Attempt to Copyright Fully AI-Generated Art Shot Down by US Supreme Court
AI Tutorial: How to Dub Your Videos Into Multiple Languages With AI
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
OPENAI
OpenAI Introduces New ChatGPT 5.3 Instant With Less "Cringe" and Fewer Hallucinations
OpenAI is updating the default and most-used model for ChatGPT users with a new version designed to address complaints about tone, relevance, and conversational flow, issues that don't always surface in standard evaluations.
Details:
The company acknowledges users had long been complaining that GPT-5.2 Instant had a "cringe" tone that came across as overbearing or made unwarranted assumptions about their intent or emotions.
The new GPT-5.3 Instant model has a more natural conversational style and cuts back on dramatic phrases like "Stop. Take a breath" or "First of all, you're not broken."
Another widely reported issue was that it would refuse questions it should have been able to answer, or respond in ways that felt overly cautious around sensitive topics. The new model cuts down on refusals, tones down overly defensive or moralizing preambles, and drops unnecessary caveats.
It also reduces hallucinations by up to 26.8% compared to its predecessor and provides higher-quality answers based on web information, as it is now better at "balancing what it finds online with its own knowledge and reasoning."
The new model is available on both ChatGPT and the API. Only the Instant model will be upgraded to 5.3, but OpenAI said it is working on updating the other models under ChatGPT, Thinking, and Pro to 5.3 soon.
TOGETHER WITH WISPR FLOW
Speak your prompts. Get better outputs.
The best AI outputs come from detailed prompts. But typing long, context-rich prompts is slow - so most people don't bother.
Wispr Flow turns your voice into clean, ready-to-paste text. Speak naturally into ChatGPT, Claude, Cursor, or any AI tool and get polished output without editing. Describe edge cases, explain context, walk through your thinking - all at the speed you talk.
Millions of people use Flow to give AI tools 10x more context in half the time. 89% of messages sent with zero edits.
Works system-wide on Mac, Windows, iPhone, and now Android (free and unlimited on Android during launch).
AI ART
Attempt to Copyright Fully AI-Generated Art Shot Down by US Supreme Court
"A Recent Entrance to Paradise," the artwork at the center of the US copyright case
The U.S. Supreme Court declined to hear an appeal regarding whether art generated entirely by AI can be copyrighted, leaving intact rulings that limit U.S. copyrights to works created by humans.
Details:
The case was brought by computer scientist Stephen Thaler, who sought to register a copyright for a 2012 image titled "A Recent Entrance to Paradise."
Thaler argued that his AI system, DABUS (Device for the Autonomous Bootstrapping of Unified Sentience), should be recognized as the author of the work, with himself as the owner of the copyright.
By refusing to hear the appeal, the Supreme Court left intact a 2025 decision from the U.S. Court of Appeals for the District of Columbia Circuit, which affirmed that the U.S. Copyright Office correctly rejected Thaler's application because the work lacked a human creator.
The decision seems to end the long saga over whether an AI can be registered as the author of an artwork. The message from the courts and administrative bodies is consistent: if you want IP protection, there must be significant human involvement. AI, no matter how sophisticated, is considered a tool, not a creator.
AI Tutorial
How to Dub Your Videos Into Multiple Languages With AI
Go to the HeyGen website and sign in.
Click on Translate and select "Video Dubbing."
Upload your video, or paste a YouTube or Google Drive URL, then click Next.
Select the target language.
You can also click the Advanced tab for more options like removing background noise, turning lip sync on or off, or adding captions.
Once you're done, click Translate and wait a few seconds to receive your video.
AI Tools to check out
📽 Veed.io: All-in-one video editing tool with a range of AI-powered features.
💻 GenPPT AI: AI PPT generator, powered by Claude Opus 4.6.
👉 Anakin: Convert websites into clean data at scale.
✅ Vesso: Turn raw copy into polished designs in one click.
📹 Storyship: Create professional product demo videos from screen recordings.
TOGETHER WITH WISPR FLOW
4x faster communication. Zero quality tradeoff.
Most people spend hours every day typing messages they could say in minutes. Wispr Flow fixes that.
Flow turns your voice into clean, polished text inside any app. Speak like you would to a colleague - tangents and all - and get professional output ready to send. 89% of messages go out with zero edits.
Use it for:
Email and Slack responses in seconds
Meeting follow-ups and project updates
Client communication on the go
Long-form writing without staring at a blank page
Millions of people use Flow daily, including teams at OpenAI, Vercel, and Clay. Works on Mac, Windows, iPhone, and now Android -- free and unlimited on Android during launch.
AI Findings/Resources
🤖 Xiaomi trials humanoid robots in its EV factory — says they’re like ‘interns’
😅 India's top court angry after junior judge cites fake AI-generated orders
🙈 Scientists make a pocket-sized AI brain with help from monkey neurons
⌛ AI makes history lessons much more interesting when you're walking through a historic scene with a guide
The latest in AI and Tech
Privacy technology company Deveillance announced Spectre I, the first smart device designed to stop unwanted audio recordings
The device scans its immediate environment for nearby microphones, including those in smartphones, smart speakers, and AI voice recorders, and sends out AI-generated cancellation signals that make your speech unintelligible to them.
The company is on track to generate annual revenue of almost $20 billion, more than doubling its run rate from late last year.
The new model is the fastest and most cost-efficient in the Gemini 3 series. It achieves an Elo score of 1432 on the Arena.ai Leaderboard and outperforms other models of similar tier across reasoning and multimodal understanding benchmarks.
Meta Platforms' AI-powered smart glasses are facing privacy concerns after two Swedish newspapers reported that contractors reviewing user data have accessed highly intimate recordings captured through the device.
The series consists of four compact, dense models built on the same unified Qwen 3.5 architecture:
Qwen3.5-0.8B & 2B: Optimized for edge devices and ultra-fast inference.
Qwen3.5-4B: A strong multimodal base for lightweight agents.
Qwen3.5-9B: A compact reasoning model with performance that rivals much larger models.
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 👩🏻💻