5 Free OpenAI Courses For Academic Success

In partnership with

In Today’s Newsletter

1. 5 Free OpenAI Courses For Academic Success

2. AI & Tech News

5 Free OpenAI Courses For Academic Success

• Mastering Prompts: The Key to Getting What You Need from ChatGPT 

• AI for Academic Success: Research, Writing, and Studying Made Easier 

• Organization and Automation: Managing Time and Tasks with AI 

• AI Career Prep: Resumes and Interviews 

• Collaborating with AI: Group Work and Projects Simplified 

Presented By The Superhuman AI

Find out why 1M+ professionals read Superhuman AI daily.

In 2 years you will be working for AI

Or an AI will be working for you

Here's how you can future-proof yourself:

1. Join the Superhuman AI newsletter – read by 1M+ people at top companies

2. Master AI tools, tutorials, and news in just 3 minutes a day

3. Become 10X more productive using AI

Join 1,000,000+ pros at companies like Google, Meta, and Amazon that are using AI to get ahead.

Sign up and start learning AI

AI & Tech News

ChatGPT Gets Smarter for Shopping (Source)

OpenAI has upgraded ChatGPT to show product images, prices, and ratings when you search for things to buy.

Now, if you ask for something like “best espresso machines under $200,” you’ll see product cards with photos, star ratings, and prices. Click a card, and you’ll get more info from sites like Amazon and Reddit — plus an option to ask ChatGPT questions about the product.

There are no ads, and results are based on real-time data (though OpenAI hasn’t shared its partners). But some info might still be off, so it’s smart to double-check.

The update is rolling out to all users — free and paid — starting today. You can also now chat with ChatGPT through WhatsApp by messaging or calling 1-800-CHATGPT.

Alibaba Releases Qwen3 – A Powerful Open-Source AI Model (Source)

Alibaba launched Qwen3, a new open-source AI model series that beats OpenAI o1 and DeepSeek R1 in key benchmarks. The top model, Qwen3-235B, rivals Google Gemini 2.5 Pro.

• 8 models (2 MoE, 6 dense), trained on 36T tokens

• Supports 119 languages

• Offers hybrid reasoning with a /think mode

• Available on Hugging Face, GitHub, and Qwen Chat

• Apache 2.0 license – free for commercial use

• Easy to deploy with OpenAI-compatible APIs

• Designed for both local use and enterprise-grade tasks

DeepSeek R2 Is Coming — And It Might Shake the AI World Again (Source)

After shocking markets with DeepSeek R1, China’s DeepSeek is reportedly launching DeepSeek R2, a 1.2 trillion-parameter hybrid MoE reasoning model.

• Trained on Huawei chips, not Nvidia

• 97.3% cheaper to train and run than GPT-4

• Multimodal support with advanced vision

• Trained on 5.2 PB of expert-level data (finance, law, patents)

• $0.07 per million input tokens, $0.27 per million output tokens

• Local supply chain = faster, independent AI infrastructure

• Rumored release: early May 2025

Thanks for your time and consideration!

Shailesh