Welcome to Frozen Light AI News Weekly Recap!

The only show where OpenAI just became your favorite professor, ChatGPT remembers more than your best friend, and Meta’s Llama rolled in with less filter and a lot more fire. This week, AI schooled us, tracked us (just a bit), and made its way into government offices. We've got the biggest moves, the boldest upgrades, and everything shaking up the world of artificial intelligence.

Let's dive into the latest updates:

🧠 OpenAI's ChatGPT Gets a Memory Upgrade

OpenAI has enhanced ChatGPT's memory capabilities, allowing it to recall user preferences across sessions. Whether it's your writing style or your dislike for pineapple on pizza, ChatGPT remembers. This feature is rolling out to Plus and Pro users, with options to manage or disable memory settings. 

🦙 Meta Introduces Llama 4

Meta's Llama 4 is here, aiming to be more responsive and less biased. With a refusal rate of only 2% for sensitive questions, compared to Llama 3's 7%, it's designed to handle complex queries more effectively. Meta plans to integrate Llama 4 across its platforms, including WhatsApp and Instagram. 

🔄Google Enhances Gemini Live API

Meanwhile, Google's Gemini Live API has been upgraded for real-time interactions. The new features include improved visual processing, longer memory, and the ability to handle interruptions during conversations. This makes AI interactions feel more natural and human-like. 

🎓 OpenAI Launches AI Academy

One of the biggest drops this week? The launch of OpenAI Academy, a free, accessible platform aimed at making AI education more democratic and less daunting. It’s designed to serve everyone from curious beginners to educators and small business owners looking to level up their AI literacy.

But it didn’t just make headlines - it made waves in our community.

Our experts jumped in with their thoughts:

🧠 Gal Miller sees the move as inevitable and transformative. In "A Sign of the Times", he positions the academy as a cultural milestone in AI's mainstream adoption, calling it “the beginning of the end for AI elitism,” and comparing it to the early days of the internet.

📚 Jeremiah Johnson, in "A New Era for LLM Education", highlights how the academy could redefine how language model education is delivered. He praises the move but urges OpenAI to invest in real pedagogical thinking to avoid simply repackaging tutorials as education.

🎯 Sarit Lahav brings a critical eye in "Teaching or Tuning?". She questions whether OpenAI’s academy is about empowering users—or subtly nudging them into their ecosystem. Her perspective leans skeptical, asking whether this is education or just clever product onboarding.

🎓 Galit Feige, in "A Nice Start, But Don’t Drop Your AI Course Just Yet", appreciates the accessibility but warns not to overhype it. She argues that structured academic programs still offer depth and rigor the academy can't yet match.

🧘‍♂️ Avinoam Boaron takes a philosophical turn in "Has OpenAI Academy Just Killed the AI Guru?". He reflects on the democratization of AI knowledge and wonders if the age of the "AI expert" is over, replaced by self-service wisdom at scale.

🧭 Yael Raz delivers practical praise in "Finally, One Place to Learn AI Without Getting Lost". She celebrates the simplicity and clarity of the academy’s interface, noting how overwhelming AI education can feel elsewhere.

🌊 Asaf Rozanes wraps it all in a big-picture question in "Navigating the Learning Curve or Just Riding the Wave?". He sees potential but challenges OpenAI to go beyond trendy optics and build a real long-term learning framework.

🏛️ White House Appoints Chief AI Officers

In a political plot twist,  te White House has mandated that all federal agencies appoint Chief AI Officers. This move aims to integrate AI into government operations, focusing on innovation and responsible use. It's a significant step in embracing AI at the federal level. 

Jordan Ledwein sees this as more than a government shift — it's a signal to the private sector. He argues that a Chief AI Officer isn’t just a tech figure, but a strategic leader who drives real transformation by aligning AI with business goals, change management, and responsible innovation.

🎶YouTube Introduces AI Music Assistant

Last but not least, YouTube's new AI Music Assistant allows creators to generate custom, copyright-free background music using text prompts. This tool simplifies the video creation process, making it easier for content creators to enhance their videos with suitable soundtracks. 

That's a wrap for this week's AI news!

See you next week with more updates from the world of AI.

👉 In the meantime, join our community to stay ahead of the curve.
You don’t want to miss what’s coming next.

 

Share Article

Get stories direct to your inbox

We’ll never share your details. View our Privacy Policy for more info.