Why you can't trust AI experts anymore
Inside: ClickUp unveils Brain 4.0, a warning on India's data privacy, and why your AI calendar is lying to you.
Welcome to issue #89 of FutureBrief. Three times a week I share practical insights on AI & automation trends, tools, and tutorials for business leaders. If you need support on your technological journey, join our community and get access to group chat, Q&As, workshops, and templates.
Ninjabot, a ready-to-deploy AI agent that responds to inquiries in 10-60 seconds (not hours), reaches out to relevant contacts on social media 24/7 (no spam), and warms up your database via SMS/email (including follow-ups).
🔮 Today’s insights
Privacy leaks exposed in India’s AI boom
A Stanford study reveals that major US tech companies are feeding user inputs back into models to train them, often without clear consent. This is happening largely in India due to a regulatory vacuum compared to the EU or US. If you outsource operations to India, your data might be training public models. Audit your BPO contracts immediately to ensure zero retention policies are actually active.
The Paper Mill scandal rocks academia
A single student, Kevin Zhu, published 113 AI-generated research papers in one year. This vibe coding approach to science produced content that looked plausible but was scientifically shallow, highlighting how easy it is to fake expertise. Credentials are becoming unreliable. When hiring AI consultants, ignore the portfolio volume. Ask for proof of their process to ensure they actually understand what they are building.
Aha 2.0 automates influencer management
Aha 2.0 launched as an AI Employee that handles the entire influencer marketing lifecycle. It discovers talent, negotiates prices, manages contracts, and even holds payments in escrow until work is verified. This democratizes high-level marketing. You can now run a complex influencer campaign without an agency, as the software handles the messy negotiation and admin work.
Notion AI struggles with calendar math
Users are reporting that Notion AI is consistently misidentifying days of the week for future dates (e.g., getting the day wrong for December 2025 dates). This is a reminder that LLMs are probabilistic text predictors, not calculators. Don’t trust a chat model for deterministic planning without verification.
ClickUp unveils Brain 4.0 and Agents
ClickUp released version 4.0, featuring a major redesign and two new AI agents powered by their Qatalog acquisition. The “Contextual Omnivore” brain can now connect disparate data sources (like Figma files and GitHub repos) to answer questions about project status without manual updates. This moves us closer to the “self-updating” AI project management. Instead of nagging your team for updates, the AI checks the file activity and updates the task for you.
💡 Vibe coding is creating a lemon market for expertise
We are entering the era of vibe coding. This is where people use AI to generate code, content, or strategy that feels right and looks professional but lacks structural integrity. A dangerous trend that the Kevin Zhu 113 papers scandal perfectly illustrates.
In economics, a lemon market happens when the buyer cannot distinguish between a high-quality product and a low-quality one (a lemon). When this happens, trust evaporates, and the market collapses.
I see this with clients hiring AI experts or developers. Candidates present impressive portfolios generated mostly by ChatGPT, but they crumble when asked to debug a complex scenario or explain the architecture. They are vibe coders. They know how to prompt, not how to build.
When hiring for your business, stop looking at the finished product. The finished product is now cheap. Look at the process. Ask candidates to walk you through how they solved a problem, not just what they built.
If they can’t explain the logic, they are just renting intelligence, not owning it.
🔧 Tools & Resources
CyberCut AI - Video Repurposing, best for turning long-form webinars or podcasts into viral clips for TikTok/Reels automatically.
Aha 2.0 - AI Influencer Manager, best for automating the entire influencer marketing lifecycle, from discovery to contract negotiation.
ClickUp Brain 4.0 - Contextual AI, best for connecting project data across apps (GitHub, Figma) to automate status reporting.
💬 Community Q&A
Q: “Notion AI keeps getting dates wrong. Why?”
A: AI predicts the next word, it doesn’t do math. When you ask “What day is Dec 25, 2025?”, it guesses based on patterns, it doesn’t look at a calendar database. For scheduling, use predefined deterministic tools or scenario functions (like in n8n or Make) rather than asking an LLM to guess the date. Always separate “reasoning” (AI) from “calculation” (code).
🎓 Next workshop
Build Your AI Research Assistant in 60 Minutes
Replace 10 hours of weekly research with an AI system for competitor analysis and market trends. Premium members build it live with Yuri.
Forward this to a colleague who’s wrestling with manual processes. They’ll thank you.
What’s your take on today’s topics? Did you like it, or is there something I missed?
Building modern tech for SMBs? Reach 20,000+ decision-makers who are actively implementing AI, automation, and no-code tools. Become a sponsor
Really sharp take on the lemon market dynamic here. The Kevin Zhu case perfectly illustrates how easy it is to fake credentialsin this space now. I run into this constantly when evaluating vendors, portfolios look impressive but then they cant explain why they chose a specific architecture over another. The process vs product framing is exaclty right, thats the only way to separate folks who understand the system from those just renting intelligence.