Between the Analog and the Algorithm

From agents to the fight for trust online and the changing creator ecosystem, three signals shaping the next phase of AI.

This week I found myself moving between two very different worlds: deep yoga teacher training and the rapidly evolving world of AI builders.

I’m knee-deep in my yoga teacher training right now — writing long scripts and practicing daily — while still juggling my creative tech life. The contrast has been mind-bending: moving from slow, analog presence to the fast-moving world of AI and builders.

Earlier this week I made it to the DeepStation meetup in Downtown Miami, and the turnout was fantastic. What stood out most was how many people in the room weren’t engineers — marketers, founders, creatives — all building and bringing their ideas to life at surprising speed.

Around the same time, I also came across an incredible list of women building in AI that I need to sit with and fully explore. (Or maybe I should just build an agent to digest it for me.)

Meanwhile, I’ve been continuing to experiment with Lovable, tweaking my weekly digest so it becomes more powerful and robust. It’s honestly wild how easily you can now use natural language to make things appear. Sometimes it genuinely feels like magic.

To help make sense of everything happening in AI each week, I’ve started running the biggest stories through a small tool I built to surface the deeper patterns behind the headlines.

Here are the three signals that stood out this week.

📊 The 3 AI Signals That Matter This Week

1. From Generative to Agentic: AI Gets a Job

A massive transition is underway from AI being a ‘chat box’ to being a ‘co-worker.’ The launch of GPT-5.4 with native computer use and financial plugins, alongside the acquisition of Forethought for ‘agentic’ customer service, signals that AI is moving into execution mode. It is no longer just generating text; it is interacting with software, managing workflows, and performing autonomous tasks. This ‘harness engineering’ separates gimmicky AI from production-ready tools.

📰 Supporting stories:

• Zendesk Acquires AI-Driven Customer Service Pioneer Forethought

• Perplexity’s Personal Computer turns your spare Mac into an AI agent

• OpenAI’s Sora Video Generator Reportedly Coming To ChatGPT

• OpenAI Buys Promptfoo to Enhance AI Agent Security

• OpenAI’s New GPT-5.4 Signals Progress Toward Autonomous AI Agents

🔮 Prediction: By the end of 2025, the ‘AI Prompt Engineer’ role will be replaced by ‘AI Agent Orchestrators’—professionals who don’t just write prompts, but manage a fleet of autonomous agents that handle end-to-end business processes like media buying, data analysis, and multi-platform content distribution.

❓ Ask yourself: If an AI agent can execute 80% of your current technical workflow (filing, data entry, posting), what is the unique human ‘judgment’ you provide that justifies your rate?

2. The Verification War: Fighting the ‘Year of AI Slop’

As AI-generated content (and ‘AI Slop’) floods the internet, platforms and regulatory bodies are racing to re-establish trust. We are seeing a 360-degree push for transparency: Apple Music and Meta are implementing AI labels, while news organizations are partnering with startups to detect misinformation. Simultaneously, users are retreating to ‘analog’ trust signals, like peer recommendations and human-centric networking, because they no longer trust automated outputs or ‘inauthentic’ brand stunts like McDonald’s ‘Burgergate.’

📰 Supporting stories:

· NewsGuard and Pangram Partner to Combat AI-Generated Misinformation in News

· B2B Buyers Prioritize Peer Trust Over AI Chatbots

· Apple Music Introduces Transparency Tags for AI-Generated Content

· AI Makes Human Judgment More Valuable, Not Less

· Meta’s Deepfake Moderation Under Fire from Oversight Board

🔮 Prediction: ‘Verified Human’ will become a more valuable premium brand tier than any ‘AI-Powered’ feature. Brands that can prove a human-in-the-loop process will command 20-30% higher price points in the creator economy.

❓ Ask yourself: Are you building ‘trust signals’ that an AI can’t fake, or are you still relying on metrics that a bot can easily replicate?

3. The Great Ecosystem Collapse: Creators vs. Institutions

The traditional boundaries of where and how we consume content are dissolving. Long-form creators are successfully ‘invading’ the living room TV, challenging big-budget publishers for Gen Z’s attention. Meanwhile, retail giants like Walmart are buying hardware companies (Vizio) to turn your TV into a direct-purchase terminal. This ‘everything-is-everywhere’ model—including the ‘fediverse’ expansion of app stores—means creators must think beyond platform silos and focus on ecosystem-wide visibility.

📰 Supporting stories:

· Long-Form Creators Eyeing TV Takeover, Chasing Bigger Brand Budgets

· Walmart’s Vizio Acquisition Expected to Unleash Retail Advertising Growth

· Generational Gap Grows: Publishers Losing Young Audiences To Creators

· Alternative App Store AltStore PAL Joins The Fediverse

· Peacock Flocks to AI-Powered Video and Mobile-First Live Sports

🔮 Prediction: The ‘Streaming Wars’ are over; the ‘Ecosystem Wars’ have begun. Success will no longer be measured by followers on one app, but by the ability to move a fan from a TikTok short to a YouTube long-form video on a TV, and finally to a ‘shoppable’ transaction powered by a retail media network.

❓ Ask yourself: Is your content strategy optimized for a ‘mobile-first’ world that ends in a ‘TV-sized’ purchase?

See you next week! — Karina