• Horizon AI
  • Posts
  • New Year’s Resolutions Fail Without a Plan. Use These Prompts to Make Them Stick ✍️

New Year’s Resolutions Fail Without a Plan. Use These Prompts to Make Them Stick ✍️

Yann LeCun Launches New ‘World Model’ AI Startup 🌎

In partnership with

Welcome to another edition of Horizon AI,

Every year starts with big intentions, but most resolutions fail because there’s no clear plan behind them. Wanting to “get fit,” “save money,” or “learn something new” sounds good, but it’s hard to follow without structure. In today’s issue, we are sharing 10 prompts that will help you turn your big ambitions into a concrete to-do list for 2026.

Let’s jump into it!

Read Time: 4.5 min

Here's what's new today in the Horizon AI

  • Yann LeCun Confirms New ‘World Model’ AI Startup

  • Meta Is Developing New Image and Video AI Models for 2026

  • AI Tutorial: 10 Gemini prompts to stick to your New Year’s resolutions

  • AI Tools to check out

  • AI Findings/Resources

  • The Latest in AI and Tech 💡

AI News

AMI LABS

Yann LeCun Confirms New ‘World Model’ AI Startup

Yann LeCun, a Turing Award winner and former chief AI scientist at Meta, has confirmed the launch of a new startup focused on “world model” AI, marking his latest move beyond LLMs.

Details:

  • The new company is called Advanced Machine Intelligence (AMI Labs), with LeCun serving as Executive Chairman rather than CEO.

  • Alex LeBrun, co-founder and CEO of medical AI startup Nabla, has been appointed CEO of AMI Labs, a move confirmed by both Nabla and LeCun.

  • The startup focuses on world model AI, an approach in which systems seek to understand their environment in order to simulate cause-and-effect and what-if scenarios to predict outcomes, positioning it as an alternative to traditional LLMs, which can hallucinate due to their “non-deterministic” nature.

  • AMI Labs is reportedly seeking to raise around €500 million at a valuation near €3 billion, even before publicly launching a product.

  • Nabla confirmed it will search for a new CEO, and partner with AMI Labs to use its models as they are developed.

This marks yet another sign of world models gaining traction across the industry, with major players such as Google DeepMind and startups like Fei-Fei Li’s World Labs pursuing similar approaches.

TOGETHER WITH GLADLY

Smarter CX insights for investors and founders

Join The Gladly Brief for insights on how AI, satisfaction, and loyalty intersect to shape modern business outcomes. Subscribe now to see how Gladly is redefining customer experience as an engine of growth—not a cost center.

META

Meta Is Developing New Image and Video AI Models for 2026

Meta is building a new generation of AI models under its superintelligence lab, led by Scale AI co-founder Alexandr Wang, as the company looks to regain momentum in the AI race.

Details:

  • Meta is developing an image and video model codenamed ‘Mango,’ alongside a new text-based model known internally as ‘Avocado’ that focuses heavily on stronger coding capabilities, while also exploring the development of world models.

  • The models are planned for release in the first half of 2026, based on an internal roadmap shared by Wang and Meta’s chief product officer, Chris Cox.

  • Meta has undergone major AI restructuring this year, including leadership changes and aggressive hiring from rival labs, though several high-profile hires have already exited.

The company is currently trailing competitors like OpenAI, and Google, with no clear breakout AI product yet; as a result, the early outputs from Meta Superintelligence Labs will be closely watched, as a key indicator of Meta’s position in advanced AI.

AI Tutorial

10 Gemini prompts to stick to your New Year’s resolutions

Big goals fail when there’s no plan. Gemini can help you turn vague resolutions into clear actions, schedules, and habits you can actually follow.

  1. Build a workout plan

Tell Gemini your current level, goal, and schedule. It creates a step-by-step plan that ramps up over time.

Prompt: “I’m planning to run the New York City Marathon in six months. I recently completed a half marathon in two hours and can train five days per week. Create a month-by-month training plan to steadily improve my endurance.”

  1. Declutter without stress

Break a big cleanup into short daily tasks so it doesn’t feel overwhelming. You can use Gemini Live with camera sharing to get real-time feedback.

Prompt: “I want to organize and declutter my apartment this month. Create a 30-day plan that breaks the process into simple, daily tasks that take no more than 20 minutes each and won’t require me to purchase any new items.”

  1. Learn a new language

Gemini can structure a learning path and create interactive study tools to test your progress.

Prompt: “I want to learn basic French. Create a 30-day guided study plan, a multiple-choice quiz focused on verb conjugations, and flashcards for 50 essential phrases.”

  1. Level up your career skills

Use Gemini to practice interviews, review presentation recordings, and even help enhance your professional image. For an extra boost, upload a headshot and refine it for platforms like LinkedIn.

Prompt: “I have a job interview next week for a [role]. Run a realistic, conversational mock interview focused on behavioral questions. Then, after I upload my practice video, provide detailed feedback on pacing, body language, and filler words.”

  1. Eat healthier

Ask Gemini to plan meals around your dietary preferences and schedule, removing the guesswork from weeknight cooking.

Prompt: “I want to eat healthier. Every Sunday at 9 a.m., generate a 7-day dinner plan that’s high-protein and vegetarian-friendly. I have about 45 minutes to cook each night. Include a weekly grocery list.”

As you build momentum, you can refine your plans, ask Gemini to add tasks to your calendar to stay organized, and even create your own Gems to remember your goals and act as dedicated experts.

AI Tools to check out

🎮 Gambo AI: The first vibe-coding game builder. Type an idea and it builds a playable game you can launch instantly.

🌌 Nimo: End your AI app chaos with a canvas that connects your everyday apps and tools.

🐟 Fish Audio: Voice generation with emotion control, voice cloning, and pro audio tools.

💼 Resumly: Automates your job search with AI-powered job matching, resume creation, and one-click apply.

 Antigravity: Google’s agentic development platform, evolving the IDE into the agent-first era.

TOGETHER WITH BEEHIIV

You can (easily) launch a newsletter too

This newsletter you couldn’t wait to open? It runs on beehiiv — the absolute best platform for email newsletters.

Our editor makes your content look like Picasso in the inbox. Your website? Beautiful and ready to capture subscribers on day one.

And when it’s time to monetize, you don’t need to duct-tape a dozen tools together. Paid subscriptions, referrals, and a (super easy-to-use) global ad network — it’s all built in.

beehiiv isn’t just the best choice. It’s the only choice that makes sense.

AI Findings/Resources

📚 Stanford CS230: Career Advice in AI

😱 Google was at risk of losing its dominance — until it promoted this AI executive

The latest in AI and Tech

Qwen has introduced Qwen-Image-Layered, a new image editing model that splits photos into separate, editable RGBA layers, similar to working in Photoshop. Each layer can be edited independently, making it easy to resize, recolor, move, or delete objects without affecting the rest of the image.

AI coding assistant Cursor has acquired Graphite, a startup focused on AI-powered code review and debugging. While deal terms weren’t disclosed, Cursor reportedly paid well above Graphite’s last $290 million valuation.

Alphabet has agreed to buy data center and energy infrastructure provider Intersect for $4.75 billion in cash, plus assumed debt, as it scales computing and power capacity to support its AI efforts.

Kuaishou has rolled out new updates to its Kling 2.6 video model, it can now generate spoken dialogue, narration, singing, and ambient sounds synced to video, and supports custom voice uploads for consistent characters. Motion control has also improved, with better full-body movement, clearer hand actions, and more natural facial expressions and lip sync.

That’s a wrap!

Thanks for sticking with us to the end! Let’s stay connected on LinkedIn and Twitter.

We'd love to hear your thoughts on today's email!

Your feedback helps us improve our content

Login or Subscribe to participate in polls.

Not subscribed yet? Sign up here and send it to a colleague or friend!

See you in our next edition!

Gina 👩🏻‍💻