In partnership with
The Future of Shopping? AI + Actual Humans.
AI has changed how consumers shop by speeding up research. But one thing hasn’t changed: shoppers still trust people more than AI.
Levanta’s new Affiliate 3.0 Consumer Report reveals a major shift in how shoppers blend AI tools with human influence. Consumers use AI to explore options, but when it comes time to buy, they still turn to creators, communities, and real experiences to validate their decisions.
The data shows:
Only 10% of shoppers buy through AI-recommended links
87% discover products through creators, blogs, or communities they trust
Human sources like reviews and creators rank higher in trust than AI recommendations
The most effective brands are combining AI discovery with authentic human influence to drive measurable conversions.
Affiliate marketing isn’t being replaced by AI, it’s being amplified by it.
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 published with the intention of making AI news and technology more accessible to everyone.
THE FRONT PAGE
Security Company's AI Surveillance Cameras Had No Security At All
TLDR: Flock Safety left 60+ AI-powered cameras streaming to the open internet without passwords, letting anyone watch people at playgrounds, parking lots, and bike paths in real time.
The Story: Security researchers discovered dozens of Flock's Condor surveillance cameras broadcasting live feeds without requiring any login credentials. Unlike Flock's license plate readers, these cameras use AI to automatically zoom in on people's faces as they walk by—tracking a rollerblader down a Georgia bike path, following a woman walking her dog, filming kids on playground swings. 404 Media verified the exposure by having a reporter stand in front of cameras in Bakersfield while watching himself on the livestream from hundreds of miles away. The feeds included 30 days of archived footage anyone could download, plus administrative controls to change camera settings. Flock told 404 Media it was a "limited misconfiguration" that's been fixed, but didn't explain what caused it or how long the cameras were exposed.
Its Significance: The exposure shows how AI features are turning basic security cameras into sophisticated people-tracking systems that cities are deploying without always understanding the security risks. Flock operates thousands of cameras across the U.S., often selling them to police departments as part of broader surveillance networks that combine license plate readers, gunshot detection, and now AI-powered face tracking. When surveillance infrastructure itself becomes a security vulnerability, it raises questions about whether law enforcement agencies are equipped to protect the sensitive data they're collecting on everyone who walks past these cameras.
QUICK TAKES
The story: Italy told Meta to stop its new rule that blocks other AI chatbots like ChatGPT and Claude from working on WhatsApp. Meta planned to ban these chatbots starting in January so only its own Meta AI would be available. Italy's competition authority said this hurts other companies and limits what users can choose. The European Commission is also looking into the rule.
Your takeaway: This is one of the first big fights over who gets to put AI assistants on the apps we use every day. If Italy wins, it could open the door for many AI companies to reach WhatsApp's billions of users.
The story: A group of writers, including "Bad Blood" author John Carreyrou, is suing Anthropic, Google, OpenAI, Meta, xAI, and Perplexity. They say these companies used stolen copies of their books to train AI systems. An earlier lawsuit led to a $1.5 billion settlement, but some authors think that deal pays too little and lets AI companies off the hook.
Your takeaway: Writers are pushing back harder against how AI companies use their work. This new lawsuit could force bigger payouts or change how AI systems are trained.
The story: When a big power outage hit San Francisco last weekend, about one-third of the city lost electricity. Traffic lights went dark across major roads. Waymo's self-driving cars kept running and passed through more than 7,000 dark signals. But the cars sometimes paused to double-check before going through intersections, which slowed things down. Waymo pulled its cars off the road when the city asked people to stay home so emergency crews could work.
Your takeaway: Self-driving cars can handle tricky situations like power outages, but they're not perfect yet. Waymo is now updating its system so cars can move more smoothly when traffic lights go dark.
The story: A new service called SayIntentions.AI brings smart air traffic control to Microsoft Flight Simulator. The AI controllers talk to pilots using natural speech and can handle both VFR and IFR flights. The system has over 1,500 different pilot voices and 800 controller voices, some with local accents. It also adds AI airline traffic that follows real flight schedules, so airports feel busy and alive.
Your takeaway: Flight sim fans can now practice radio calls anytime, even when human controllers aren't online. This makes home simulators feel much more like real flying.
TOOLS ON OUR RADAR
🔇 Krisp
Freemium: Remove background noise from both sides of your calls with AI that works with any conferencing app—eliminates barking dogs, crying babies, and keyboard clicks instantly.📹 Whereby
Freemium: Host video meetings in your browser with no downloads required—create your own personal meeting room with a custom URL that's always ready to use.📊 Plausible Analytics
Paid: Track website traffic with privacy-first analytics that doesn't use cookies or collect personal data—simple, lightweight alternative to Google Analytics.🦊 Mullvad Browser
Free and Open Source: Browse privately with this hardened Firefox built by VPN experts—blocks trackers and fingerprinting by default without requiring an account or even a Mullvad VPN subscription.
TRENDING
AI Agents Want Full Access to Your Data — AI assistants that can browse the web and do tasks for you need access to your emails, calendar, and messages to work well. Privacy experts warn this gives tech companies even more of your personal information than before.
Amazon's Alexa+ Adds Travel and Local Services — Alexa+ now works with Expedia, Yelp, Angi, and Square, letting users book hotels, find restaurants, and schedule appointments just by talking.
Woman Describes AI Image Obsession That Led to Mental Health Crisis — A startup worker says spending hours each day making AI images of herself triggered a manic episode. She became obsessed with looking like her AI-generated photos and almost hurt herself before getting help.
Startup Raises $10.5M for AI Video Avatars — Lemon Slice can turn a single photo into a talking video avatar that streams at 20 frames per second. Companies can use it for customer service, tutoring, or training videos.
Game Loses Indie Awards After AI Use Revealed — Clair Obscur: Expedition 33 had its Game of the Year and Debut Game awards taken away after the studio admitted it used AI to make placeholder images during development. The awards went to runner-ups Blue Prince and Sorry We're Closed.
Google's Disco Turns Browser Tabs Into Mini Apps — A new Chrome experiment called Disco lets users describe what they want, and the AI builds a custom app from their open tabs. It works for things like study guides or travel planning.
TRY THIS PROMPT (copy and paste into ChatGPT, Claude, or Gemini)
Cognitive Distortion Identifier: Catch unhelpful thinking patterns and reframe them using CBT-based techniques
Build me an interactive Cognitive Distortion Identifier as a React artifact that helps recognize and challenge negative thought patterns using cognitive behavioral therapy principles.
The console should include these sections:
1. **Thought Capture** - Record what's bothering you:
• Large text area: "What's going through your mind?"
• Situation context (what triggered this thought)
• Emotion selector with intensity (1-10):
- Anxious, Sad, Angry, Guilty, Ashamed, Frustrated, Hopeless
• Physical sensations (tight chest, racing heart, etc.)
• Time/date stamp
• "Analyze this thought" button
2. **Distortion Detector** - Identify patterns:
• AI scans your thought for 15+ common distortions:
**All-or-Nothing Thinking**: Black and white, no middle ground
- Example: "I failed, so I'm a complete failure"
**Overgeneralization**: One event = always/never pattern
- Example: "This always happens to me"
**Mental Filter**: Only seeing the negative
- Example: Ignoring 9 compliments, focusing on 1 criticism
**Discounting the Positive**: Good things don't count
- Example: "That success was just luck"
**Jumping to Conclusions**: Mind reading or fortune telling
- Example: "They think I'm incompetent" / "This will fail"
**Catastrophizing**: Imagining worst-case scenarios
- Example: "If I mess up, everything will fall apart"
**Emotional Reasoning**: Feeling = fact
- Example: "I feel stupid, therefore I am stupid"
**Should Statements**: Rigid rules for self/others
- Example: "I should always be perfect"
**Labeling**: Global judgments from single events
- Example: "I'm a loser" instead of "I made a mistake"
**Personalization**: Taking blame for external events
- Example: "It's my fault they're upset"
• Highlight distortions in your text
• Severity rating per distortion
• Most common pattern identified
3. **Evidence Examination** - Challenge the thought:
• **Evidence For**: What supports this thought?
• **Evidence Against**: What contradicts it?
• Balanced questions to prompt:
- "Is there another way to look at this?"
- "What would I tell a friend in this situation?"
- "Am I confusing a thought with a fact?"
- "What's the worst/best/most realistic outcome?"
- "Will this matter in 5 years?"
• Evidence strength meter
• Reality check score
4. **Reframe Generator** - Alternative perspectives:
• AI generates 3-5 balanced reframes:
- More accurate version
- More helpful version
- Compassionate version
• Each reframe shows:
- New thought statement
- Why it's more balanced
- How it might feel different
• "Try this reframe" selector
• Custom reframe editor
• Emotion re-rating (how do you feel now?)
5. **Pattern Tracker** - Recognize themes:
• Journal of past thoughts
• Most frequent distortions (pie chart)
• Triggering situations (what sets you off)
• Progress over time (getting better at catching them?)
• Emotional trends (which emotions link to which distortions)
• "Your top 3 patterns" summary
6. **Quick Reference** - Distortion library:
• Expandable guide for each distortion type:
- Definition
- Examples
- Why it's unhelpful
- How to challenge it
- Healthier alternative
• "Am I doing this?" self-check questions
• Printable reminder cards
7. **Coping Tools** - In-the-moment help:
• "I'm spiraling" emergency mode:
- Grounding exercise (5-4-3-2-1 senses)
- Deep breathing visualization
- Distraction techniques
- When to seek professional help
• Self-compassion prompts
• Positive coping statements
• Crisis resources (when thoughts are dangerous)
8. **Progress Dashboard** - Track improvement:
• Thoughts logged this week/month
• Average emotion intensity (trending down?)
• Reframing success rate (do new thoughts feel better?)
• Most helpful reframes (what works for you)
• Distortion frequency (catching them earlier?)
• "Mental fitness score" based on patterns
Make it look like a thoughtful mental health tool with:
• Thought bubble visual metaphor
• Calm, supportive color palette (soft blues, greens, warm neutrals)
• Gentle, non-judgmental language throughout
• Analysis overlays highlighting distortions
• Before/after thought comparison
• Journal/diary aesthetic
• Progress tracking with encouraging feedback
• Clean, uncluttered interface (reduce overwhelm)
• Accessible, warm design
• "You're not alone" supportive tone
When I click "Search CBT Resources" or "Find Mental Health Support," use web search to find cognitive behavioral therapy resources, mental health support information, and evidence-based coping strategies.What this does: Helps you catch unhelpful thinking patterns in real-time using CBT principles—identifying cognitive distortions, examining evidence, generating balanced reframes, and tracking patterns over time so you can challenge negative thoughts before they spiral.
What this looks like:
WHERE WE STAND (from today’s stories)
✅ AI Can Now: Simulate 1,500+ unique pilot voices with regional accents for flight simulator radio calls.
❌ Still Can't: Match the subtle stress and emotion in a real human controller's voice during a busy traffic rush.
✅ AI Can Now: Build custom mini-apps from your browser tabs using plain English instructions.
❌ Still Can't: Run on most computers — Google's new Disco tool only works on Macs through an invite-only test program.
✅ AI Can Now: Power voice assistants that book hotels, find plumbers, and schedule salon appointments through natural conversation.
❌ Still Can't: Complete complex multi-step tasks reliably — current AI agents often fail partway through longer requests.
FROM THE WEB
RECOMMENDED LISTENING/READING/WATCHING
Benn Jordan doesn't just expose the Flock surveillance network; he demonstrates its vulnerabilities and shows how easily the system can be fooled. It's equal parts unsettling and empowering, perfect for understanding how AI surveillance actually works in your neighborhood right now.
The framework thousands of teams use daily
Organizations worldwide are adopting Smart Brevity to move faster with clearer comms. This free 60-minute session shows you how the methodology works and how to start applying it immediately.
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.
