In partnership with
Your competitors are already automating. Here's the data.
Retail and ecommerce teams using AI for customer service are resolving 40-60% more tickets without more staff, cutting cost-per-ticket by 30%+, and handling seasonal spikes 3x faster.
But here's what separates winners from everyone else: they started with the data, not the hype.
Gladly handles the predictable volume, FAQs, routing, returns, order status, while your team focuses on customers who need a human touch. The result? Better experiences. Lower costs. Real competitive advantage. Ready to see what's possible for your business?
Beginners in AI
Good morning and thank you for joining us again!
Welcome to this daily edition of Beginners in AI, where we explore the latest trends, tools, and news in the world of AI and the tech that surrounds it. Like all editions, this is human curated and edited, and published with the intention of making AI news and technology more accessible to everyone.
THE FRONT PAGE
OpenAI Is Paying People to Upload Their Old Jobs So AI Can Learn to Do Them
TLDR: OpenAI is paying contractors to upload real deliverables from past jobs to teach AI agents how to do office work.
The Story:
OpenAI and training data company Handshake AI are asking contractors to upload actual work products from current or previous employers, including Word docs, PDFs, PowerPoints, spreadsheets, and code repos, according to internal documents obtained by WIRED. Not summaries. The actual files. The goal is to establish a "human baseline" so OpenAI can measure how well its AI models perform against real professionals on real tasks. One example from the presentation asks for a luxury travel itinerary created for an actual client. Another requests "long-term or complex work (hours or days+)" that contractors have done in their occupations. OpenAI says this benchmarking is a "key indicator" of its progress toward AGI, an AI system that outperforms humans at most economically valuable work. The company also reportedly explored buying internal data from defunct businesses, including emails and documents, but one potential seller walked away over concerns about removing personal information.
Its Significance:
Web scraping and synthetic data can't teach AI what a real client deliverable or internal strategy doc looks like, so OpenAI is going straight to the source: paying people to hand over the work itself. It's a signal of how intense the competition for high-quality training data has become and how directly the AI industry is now targeting white-collar automation. The approach raises obvious legal questions about NDAs and trade secrets, but the bigger picture is strategic. OpenAI isn't just building chatbots anymore. It's building systems meant to replace the tasks those contractors once got paid to do.
QUICK TAKES
The story: Google removed AI-generated summaries for certain medical queries after a Guardian investigation found the results failed to account for age, sex, and ethnicity when displaying liver test ranges. The company says the information wasn't necessarily wrong, but health advocates worry about AI Overviews' broader implications for medical searches.
Your takeaway: Google's AI can surface technically accurate health info that's still dangerously incomplete. One investigation fixed one query, but millions of health searches still get AI summaries with the same blind spots.
The story: IgniteTech CEO Eric Vaughan cut nearly 80% of his workforce in 2023 after employees resisted his AI-first mandate. Two years later, the company has maintained nine-figure revenue, launched multiple AI products, and can build new features in four days instead of months. Surprisingly, technical staff were the most resistant, not marketing or sales.
Your takeaway: One in three workers have actively sabotaged their company's AI rollout, according to recent research. Vaughan's extreme approach worked for him, but he doesn't recommend others follow it. The real lesson? AI adoption is a culture problem, not a tech problem.
The story: Utah partnered with startup Doctronic to let an AI agent autonomously renew 191 common medications, including drugs for cholesterol, blood pressure, and birth control. The system excludes narcotics, stimulants, and antibiotics. Safety phases start with human review of all refills, then drop to 10% sampling, then random audits.
Your takeaway: This is the first real test of AI making clinical decisions on its own in the US. Utah's regulatory sandbox lets the state suspend normal rules to gather evidence. If it works, Arizona and Texas are next. If it fails, it'll set AI healthcare back years.
TOOLS ON OUR RADAR
✍️ Documenso Free and Open Source: Sign documents and collect signatures without handing your contracts to a big tech company.
📅 SavvyCal Freemium: Let invitees overlay their calendar on yours when booking meetings, so everyone picks the best time faster.
🖊️ Junia.ai Freemium: Write SEO-optimized blog posts and articles with an AI that researches your topic and matches your brand voice.
🎬 PixVerse Freemium: Turn text prompts or images into short video clips with AI, perfect for social content and marketing.
TRENDING
Google Launches Shopping Protocol So AI Agents Can Buy Things for You - Google's Universal Commerce Protocol creates a standard way for AI agents to browse, compare, and purchase across retailers. Walmart, Shopify, Target, and Etsy helped build it, and checkout will work directly inside Google Search's AI Mode.
Wing Drone Delivery Expanding to 150 More Walmart Stores - Alphabet's drone company will reach 40 million Americans by 2027, covering LA to Miami. Top customers already order three times a week, mostly eggs, ground beef, and snacks like Takis.
Motional Reboots Robotaxi Plans With AI-First Approach - The Hyundai-Aptiv venture ditched its patchwork of individual AI models for a single foundation model. Driverless commercial service in Las Vegas is now targeted for late 2026.
Google: Don't Make Bite-Sized Content for AI if You Want to Rank - Google's Danny Sullivan warns that chopping articles into AI-friendly chunks will backfire. The company's ranking systems will evolve to reward comprehensive content written for humans, not bots.
TRY THIS PROMPT (copy and paste into Claude, ChatGPT, or Gemini)
Procrastination Pattern Breaker: Identify your triggers, break the behavior chain, and build systems that make action easier than avoidance
Build a fully functional Procrastination Pattern Breaker as an interactive React app. Render the working application immediately - no code display.
**What I procrastinate on most**: [Tasks or areas where you struggle]
**My usual avoidance tactics**: [Social media, snacks, "research," cleaning, etc.]
Create these sections:
1. **Trigger Logger**
• Log procrastination moments:
- What task were you avoiding?
- What did you do instead?
- Time of day
- How were you feeling? (tired, overwhelmed, bored, anxious, unclear)
- What triggered the avoidance?
• Pattern detection over multiple entries
• "Add Entry" button
• History view of logged moments
2. **Behavior Chain Visualizer**
• Interactive chain diagram showing:
TRIGGER → THOUGHT → FEELING → AVOIDANCE → CONSEQUENCE
• Your personal chain mapped out:
- "When I see [task]..."
- "I think [thought]..."
- "I feel [emotion]..."
- "So I do [avoidance behavior]..."
- "Which leads to [consequence]..."
• Identify the weakest link to break
• "Where can you interrupt?" selector
3. **Procrastination Type Profile**
• Quiz to identify your style:
- The Perfectionist (fear of not being good enough)
- The Overwhelmed (don't know where to start)
- The Rebel (resisting external pressure)
- The Thrill-Seeker (needs deadline pressure)
- The Dreamer (loves planning, hates doing)
• Your type result with description
• Specific strategies for your type
• Common traps for your profile
4. **If-Then Planning Builder**
• Create implementation intentions:
"IF [trigger situation], THEN I will [specific action]"
• Examples based on your logged triggers:
- "If I open Twitter, then I close it and do 5 minutes of work"
- "If I feel overwhelmed, then I write down just the next step"
- "If it's 9am, then I start my hardest task"
• Save your if-then plans
• Daily reminder of active plans
• Track which plans are working
5. **Temptation Bundling Setup**
• Pair dreaded tasks with rewards:
- [Hard task] + [Enjoyable thing]
• Suggestions based on your avoidance tactics:
- Hate email? Email + favorite playlist
- Avoid reports? Reports + coffee shop
- Skip exercise? Exercise + podcast
• Create your bundles
• "Search Temptation Bundling" for research
6. **Friction Designer**
• Add friction to distractions:
- App blockers to install
- Phone in another room
- Log out of time-wasters
- Website blockers
• Remove friction from tasks:
- Prep materials night before
- Break into tiny steps
- Set visible reminders
- Create starting rituals
• Your personal friction plan
7. **2-Minute Wins Tracker**
• List of 2-minute versions of tasks you avoid
• "Just do 2 minutes" timer
• Track streaks of starting
• Celebrate momentum wins
• "Did you keep going?" logger
Design specs:
- Behavior chain aesthetic with connected node diagrams
- Flow chart style for trigger→action sequences
- Clean analytical dashboard feel
- Warm encouraging palette (soft blues, greens, coral accents)
- Chain link graphics connecting stages
- Progress tracking with streak counters
- Interactive chain you can click to "break"
- Checkbox satisfaction animations
- Timer component for 2-minute challenges
When "Search Temptation Bundling" is clicked, use web search to find behavioral science research on habit formation and procrastination interventions.What this does: Goes beyond "just do it" advice by mapping your personal procrastination patterns. Identifies your triggers, shows where to break the behavior chain, and builds concrete systems like if-then plans and temptation bundles that make starting easier than avoiding.
What this looks like:
WHERE WE STAND(based on today’s Quick Takes and Trending news)
✅ AI Can Now: Autonomously refill your blood pressure meds in Utah, with human oversight dropping from 100% to random audits as it proves itself.
❌ Still Can't: Account for your age, sex, or ethnicity when summarizing medical test results, even when those factors completely change what "normal" means.
✅ AI Can Now: Shop across dozens of retailers using a single protocol, comparing prices and applying your loyalty discounts without you clicking through each site.
❌ Still Can't: Get workers to actually use it. One in three employees have actively sabotaged their company's AI tools, and technical staff resist more than sales or marketing.
RECOMMENDED LISTENING/READING/WATCHING
PODCAST EPISODE: Bright Spots in Healthcare - "How AI is Reshaping Healthcare - An Interview with NVIDIA's Vega Shah"
This 21-minute episode covers real-world examples of NVIDIA's AI tools enhancing clinician-patient relationships, strategies for implementing AI in healthcare organizations, and both the opportunities and challenges of adoption. A practical look at where AI is making a genuine difference in medicine today.
Thank you for reading. We’re all beginners in something. With that in mind, your questions and feedback are always welcome and I read every single email!
-James
By the way, this is the link if you liked the content and want to share with a friend.
Some links may be affiliate or referral links. This helps support the newsletter at no extra cost to you.
