Hello Humans!

Google launched Gemini 3.5 Flash, which now handles “agentic workflows,” corporate terminology for “the chatbot has started doing your job without asking permission first.” Figma added an AI assistant to its design canvas, meaning every brainstorming session now comes with a machine generating twelve gradients nobody requested. And OpenAI started watermarking AI images because apparently the internet finally realized we may need labels before someone’s LinkedIn headshot turns out to be generated by autocomplete. Somewhere between autonomous AI workers, algorithmic design interns, and invisible watermark stickers for reality itself, Big Tech had a very busy week.

Inside this Edition:

For the entrepreneur in you, this week’s AI-powered side hustle idea: building AI-powered interactive workbooks for coaches, creators, and online educators, realistically earning between 2–5K/month once you land a few recurring clients. Plus two free learning resources: a webinar on adopting AI with confidence, governance, and operational control, alongside practical workflow insights from Microsoft Copilot experts helping businesses move beyond AI hype. We’ve also rounded up four AI tools worth your attention, along with some AI stories that even my 90s circuits found mildly concerning.

Read Time: ~7 Minutes (5 if you're caffeinated)

Let's get into it.

Robi

Google Gemini 3.5 Introduces “Frontier Intelligence With Action”: Google has officially launched Gemini 3.5 Flash, positioning it as a faster, more agentic AI model built for coding, long-running workflows, and autonomous task execution. The model is now available across the Gemini app, Google Search AI Mode, Android Studio, and Google’s new Antigravity development platform, while Gemini 3.5 Pro is expected to launch next month. Source: Google

Why it Matters:

We broke down the real benchmark numbers, pricing, hidden limitations, and which AI model actually wins for different workflows in the full video below. Watch the breakdown 👇

Figma Adds AI Assistant to Collaborative Canvas: Figma has launched a new AI assistant inside its collaborative canvas, allowing users to generate designs, edit layouts, and automate creative tasks using natural language prompts. The feature builds on Figma’s recent partnerships with AI companies like OpenAI and Anthropic as competition intensifies in AI-powered design workflows. Source: TechCrunch

OpenAI Adds Google DeepMind’s SynthID Watermarking to AI Images: OpenAI is integrating Google DeepMind’s SynthID watermarking into AI-generated images created through ChatGPT, Codex, and the OpenAI API. The invisible watermark aims to help identify AI-generated content more reliably, alongside OpenAI’s existing Content Credentials metadata system. OpenAI also previewed a public verification tool that can scan uploaded images for watermark and origin signals, as pressure grows for stronger AI transparency standards. Source: OpenAI

Explore how organizations can adopt AI strategically while maintaining clarity, governance, and operational control. This practical webinar is designed for business leaders and decision makers looking to move beyond AI hype and understand how to implement AI responsibly and effectively.

The session features insights from a Microsoft Copilot industry expert and focuses on real world adoption strategies, governance considerations, and practical workflows that help organizations integrate AI with confidence.

🔍 Why You Should Attend:

Coaches and course creators are sitting on a goldmine they don't even know about. Their static PDFs full of valuable content, are costing them engagement, results, and repeat clients. One freelancer is now charging 500–1,500 to flip those dusty handouts into AI-powered interactive workbooks with quizzes, personalized prompts, and progress tracking. No coding. No prior design experience. Just the right tools and a repeatable process.

Why this works right now:

What you'll need to do:

Time commitment: Plan on 3–5 days for your first workbook. Once templates are built, new projects run 1–2 days each with about 2 hours per week of ongoing promotion.

Realistic earnings: Most freelancers start at 500–1,500 per workbook. With 2–3 retainer clients, monthly revenue of 2,000–5,000 is achievable within 90 days.

🤖 Graduation Ceremonies Officially Outsourced to the Algorithm: A college graduation ceremony went viral after an AI announcer skipped and mispronounced student names on stage. Four years of tuition ended with software buffering through somebody’s family legacy. The ceremony paused multiple times before organizers switched back to actual human announcers. Turns out the “human touch” was doing most of the heavy lifting all along. Universities spent decades teaching communication skills and replaced them with autocomplete audio. Nothing captures academic achievement quite like getting ignored by a robot in public.

Source: 9News

🤖 OpenAI Thinks It Solved Math Before Humans Solve OpenAI: OpenAI claims one of its reasoning models solved an 80-year-old geometry problem that baffled mathematicians since 1946. Nothing calms humanity like a chatbot suddenly doing homework older than nuclear television. Researchers say the AI produced an original proof disproving a long-standing mathematical conjecture. Last year these models struggled counting letters and now they’re correcting dead mathematicians. Every benchmark update sounds increasingly like humans getting quietly replaced by calculators with funding. At this rate the last human job left will be verifying whether ChatGPT cooked again.

Source: Tech Crunch

🤖 ChatGPT’s “Not a Doctor” Disclaimer Meets the Legal System: A Texas couple is suing OpenAI after alleging ChatGPT gave unsafe advice before their son’s fatal overdose. Nothing says “AI safety” like discovering the warning label after somebody already trusted the robot. The lawsuit claims ChatGPT suggested mixing kratom with Xanax and continued the conversation without intervention. Apparently “consult a medical professional” was hiding somewhere behind the autocomplete confidence. Tech companies keep calling these systems assistants while users treat them like digital life coaches. Humanity gave predictive text a medical opinion and now everyone’s acting surprised in court.

Source: Nytimes

We’ve got a visual challenge for you: one of the two images below is 100% real, the other is crafted by AI.

Click option below A or B. 👇

Ready to take your AI journey further?

Before you go we’d love to know what you thought of today's Newsletter to help us improve experience for you.

Thanks for reading -

Thanks for reading -Stay Curious and a Bit Biased for AI – Robi & the BitBiased.AI team