🤘 GPT-5.1 launches

Hey folks,

Google just rolled out podcasts in Google Drive. Drive can now convert your long, boring PDFs into conversational audio summaries complete with the kind of banter you'd expect from two hosts who actually read the thing. One click generates a 2-10 minute "podcast episode" that hits your Drive folder while you're commuting, working out, or when your boss asks if you read that 47 page document. It's the same NotebookLM magic that went viral for turning research papers into surprisingly engaging audio, except now it's baked into the tool you use every day. Finally, a legitimate reason to say "I consumed that content" instead of "I read it.”

Let’s dive in..

🖥️ GPT 5.1 released

🐝 The Buzz: OpenAI just released GPT-5.1 The company explicitly acknowledged user feedback: "We heard clearly from users that great AI should not only be smart, but also enjoyable to talk to." GPT-5.1 promises to be "warmer, more intelligent, and better at following your instructions”.

• Two-model release strategy: Instant for everyday chat, Thinking for complex reasoning and both designed to actually feel good to use.

• Response to GPT-5 criticism: Users preferred GPT-4 for math, science, and writing tasks, which is brutal when you just shipped your "next-gen" model.

• Six personality presets: Six tone presets let you dial in everything from efficient robot to quirky friend, because apparently "smart but annoying" wasn't the vibe.

💡 Takeaway: Here's the uncomfortable truth: OpenAI's internal evals probably said GPT-5 was amazing. Their users disagreed loudly. The gap between "performs well on benchmarks" and "people actually want to use this" just cost them months of development and a PR hit. The speed of the GPT-5.1 release and the explicit "we heard you" messaging, suggests OpenAI is feeling competitive pressure from Claude and Gemini on user satisfaction, not just capabilities.

Together with Synthflow

WhatsApp Business Calls, Now in Synthflow

Billions of customers already use WhatsApp to reach businesses they trust. But here’s the gap: 65% still prefer voice for urgent issues, while 40% of calls go unanswered — costing $100–$200 in lost revenue each time. That’s trust and revenue walking out the door.

With Synthflow, Voice AI Agents can now answer WhatsApp calls directly, combining support, booking, routing, and follow-ups in one conversation.

It’s not just answering calls — it’s protecting revenue and trust where your customers already are.

One channel, zero missed calls.

Try It Today

📉 Softbank Dumps Nvidia

🐝 The Buzz: When the investor who turned a $20 million Alibaba bet into $50 billion dumps his entire $5.8 billion Nvidia position, you pay attention. Masayoshi Son isn't fleeing AI, he's doubling down on it, just not with Jensen Huang's H100s. The target is Baidu's AI infrastructure, which reportedly matches GPT-5 and Gemini performance at a fraction of the cost. Son's selling at Nvidia's peak ($5 trillion market cap, fastest ascent in corporate history) to back alternative compute architectures before everyone else catches on.

• All or nothing: He sold the entire stake $5.8B stake. He’s not trimming positions, this is conviction, not portfolio rebalancing.

• Baidu's performance claims: Intelligence hitting GPT-5 and Gemini benchmarks while running on non-Nvidia infrastructure.

• Perfect timing: Exiting right as Nvidia crosses $5T suggests Son thinks the easy money's been made.

• Where the money's going: Earlier-stage AI infrastructure plays, custom ASICs, and emerging Chinese compute alternatives.

💡 Takeaway: Son's made a career out of seeing around corners like Alibaba before anyone took Chinese e-commerce seriously, ARM before mobile ate the world. If he thinks the next AI fortunes get built on something other than Nvidia GPUs, engineering leaders should at least explore the alternatives. AMD, custom ASICs, and yes, Chinese AI infrastructure aren't just cost-saving measures anymore, they're potential competitive moats.

📰 Like newsletters? Here are more our readers enjoy.

💻️ How to Create Audio Summaries In G-Drive

With large PDFs (reports, research papers, contracts) you don’t always have time to read every word. The Audio Overview feature lets you listen to a summary of the document instead.

• Go to drive.google.com

• Locate a PDF file in your Drive that you want to process

• Click to open the PDF in the Drive viewer

• At the top (right side) click the button labelled Create audio overview

• Wait a few minutes while Gemini generates the audio.

• Once finished:

  • You’ll receive an email with the link to the audio file.

  • You’ll find the audio file in your Drive under My Drive → Audio overviews.

  • You can open the side panel in Drive to listen.

🐝 AI buzz bits

🤖 Elon Musk proposed giving people free Optimus robots that follow them around to physically prevent crime, calling it "a more humane form of containment of future crime"

🧠 Yann LeCun is leaving Meta to launch a world models startup focused on spatial intelligence to build systems that understand physics and physical space rather than just language. This signals that some of AI's top researchers believe the next breakthroughs require architectural shifts beyond scaling current LLMs.

🎤 ElevenLabs launched an Iconic Voice Marketplace licensing AI clones of famous voices and people like Michael Caine and Liza Minnelli to make the first ethical solution for celebrity AI voices at scale.

🔥 AI productivity tools

For a full list of 1000+ AI tools, visit our Supertool Directory

🤘 TGIF