When AI Resets Social: Marketing’s New Meritocracy

He remembers a Sunday — a small, stubborn realization: the internet changed everything in the late 1990s, and AI is arriving with the same tectonic force. This piece frames that moment through three quick things he keeps saying in rooms: (1) AI will impact everyone like the web did in 1997–1998, (2) TikTok’s meritocracy can make a hotel-room video outsell a seasoned campaign, and (3) deepfakes force a ledger of truth — enter blockchain. The intro mixes a fleeting airport anecdote, a mild tangent about teenage curiosities, and a clear ask: pay attention, do 50–100 hours of homework, and experiment.

1) The 1997 Moment: Why AI Feels Like the Web All Over Again

Many marketers describe today’s AI wave the way they remember the internet arriving in 1997–1998: confusing at first, then suddenly unavoidable. The point is not hype. It is scale. AI is positioned as the single biggest technology shift of the last 30 years, and it will touch daily life and work at the same time.

“AI is going to impact every one of you professionally and personally the same way the internet did in 1997–1998 when it started.”

AI Trends 2026: A shift years in the making, now moving fast

In the late 1990s, the web changed how people found information, how brands built trust, and how products were sold. AI is doing something similar, but faster. It is already reshaping search visibility as answers move from blue links to AI summaries and assistants. At the same time, automation is spreading across marketing workflows, from research to reporting to content production.

That is why AI in Marketing is not a side project. It is becoming the new baseline for speed, quality, and decision-making. Research backs this up: 46% of marketers use AI to scale creative production, which means the volume of ads, posts, and variations will rise—and the competition for attention will get tougher.

Digital Marketing Trends: Homework, not headlines

The practical advice is simple: do not treat AI as a conference takeaway. Treat it as a skill. The recommended path is 50–100 hours of hands-on learning over the next six months—downloading tools, testing prompts, reading, and listening.

“You must spend 50 to 100 hours of homework — downloading stuff, playing with stuff, reading stuff, listening to podcasts.”

1. Pick 2–3 tools and use them weekly (writing, design, analytics).
2. Save prompts and results like a swipe file.
3. Follow one podcast or newsletter to track changes.

Predictive Strategy: Using history to see what comes next

A small personal aside often shared in this context: even a DNF student can spot patterns by studying history. It is imperfect, but useful. The same mindset applies now. Predictive planning is moving strategy upstream—teams can model outcomes earlier, test more ideas, and decide faster. In short, this is not incremental. It is monumental, and putting one’s head in the sand is not an option.

2) The TikTok Meritocracy: Content Wins, Followers Follow

TikTok Marketing and the rise of Interest Media

In earlier social eras, the social graph ruled. If a brand or creator had a big following, reach was more predictable. A common benchmark was that 6 million Instagram followers could almost guarantee attention on demand. Today, TikTok Marketing reflects a different rule: discovery is driven by what people are interested in right now, not who they already follow. As one line puts it, “We live in interest media. It’s showing you content of the thing you’re currently into.”

Content-first distribution: the new meritocracy

This is why TikTok feels like a meritocracy. A creator can have 15 million followers on TikTok and still see a post underperform if it does not match current viewer interest. Meanwhile, a first-time poster can win immediately. The speaker describes making a clip at the airport and having it judged on the content itself, not follower count. The point is simple: a hotel-room video can beat a big account because the algorithm rewards relevance and watch behavior.

“Your video might get more views than me.”

Real-world proof: small accounts selling out overnight

Across Digital Marketing Trends, this is one of the clearest signals of Social Platform Shifts. There are hundreds of weekly examples of someone posting about a business or product and selling out at a scale that traditional performance marketing could not reach “on its best day,” even with heavy testing around CAC, LTV, and conversion rates. The difference is distribution: TikTok can place a product story directly in front of people already primed for that category.

“That is a level of meritocracy. That is insane. That is the biggest opportunity in the world.”

An actionable test for marketers

For teams adapting their marketing stack, the practical move is to treat TikTok as an interest-led lab. AI is moving to the center of personalization at scale, but the creative still has to earn attention in the first second.

1. Post one simple video tonight (a demo, a before/after, a customer story).
2. Write for immediate interest, not brand history.
3. Measure completion rate, saves, and shares before follower growth.

Why the window may not last

This content-first advantage may change as interfaces evolve. The speaker predicts the primary device could shift in 6–7 years (for example, AR glasses), resetting distribution rules again—another reminder that Social Platform Shifts reward fast learning.

3) AI as Leverage: Scale Creative, Shrink Costs

AI is turning content production into leverage. Work that once needed large teams and long timelines can now be done by a small group with low-cost tools. As one operator put it,

“What used to cost 30 or 40 people to do something and cost millions of dollars will now be done better for $9.99 a month.”

That shift changes how Marketing Budgets are planned: less spend on manual production, more spend on distribution, testing, and strategy.

Scale Creative Without Scaling Headcount

The new advantage is volume. Instead of asking a team to post twice a day across platforms, AI makes extreme output possible. The claim is blunt:

“I’m going to post 4,000 times a day using AI.”

Whether a brand hits that number or not, the point stands: Scale Creative is no longer limited by how many editors, designers, and copywriters can be hired.

AI Generated Creative + AI Optimisation in the Same Loop

AI Generated Creative is not just about making more assets; it is about making more versions. Research shows 46% of marketers use AI to scale creative production, and the reason is simple: variation fuels performance. When AI moves to the center of the marketing stack, it can personalize at scale and support AI-assisted testing that enables real-time ad optimization. That means faster learning cycles: new hooks, new cuts, new captions, and new offers—tested continuously.

From 100 People to Five (If the System Is Right)

Efficiency is the trade. The same speaker frames it as:

“It’s going to cost me the same amount of money because it’s going to be five people using AI to do the work of what used to cost me 100 people to do.”

Many tasks—captioning, resizing, versioning, basic edits, first-draft copy, and reporting—can be handled with tools under $10/month, freeing specialists to focus on taste and direction.

Proof That Scale Compounds

VaynerMedia’s growth from zero to $350M in annual revenue over 14 years is a reminder that capability compounds when production and distribution systems improve. AI accelerates that compounding by lowering the cost of iteration.

Guardrails: Avoid “AI Slop”

• Human taste must approve outputs to prevent low-trust, automated spam.
• Use pod-based teams: small cross-functional groups (creative + media + analytics) augmented by AI.
• Measure quality signals, not just volume: saves, shares, watch time, and conversion.

4) Trust on Trial: Deepfakes, Video Proof, and the Blockchain Response

When video became the “judge and jury”

For roughly 100 years, motion pictures and video have carried a special status: they felt like reality. Public moments such as the Rodney King footage and the JFK assassination shaped how people understood truth, because viewers could see what happened. In that short slice of human history, “video proof” became the default referee for news, courts, and culture.

AI Backlash and the end of automatic belief

That assumption is now on trial. Deepfakes can generate clips that look real, sound real, and spread fast. The speaker’s warning is blunt:

“In 10 years, nobody in this conference will believe a single video they see because of AI.”

It is not a theoretical risk. He notes:

“There are videos of me right now online of saying things I never said. You would have no idea that I didn’t say it.”

This is why Deepfake detection is becoming a marketing requirement, not a niche security topic. At the same time, “AI slop” is already creating AI Backlash, pushing platforms to add AI-content labels and opt-outs. As AR and visual search reach the mainstream, authenticity concerns rise because the camera feed itself becomes a shopping and discovery interface.

Blockchain Video as a “ledger of truth”

The proposed response is provenance: proving where a video came from and whether it was altered. The speaker frames it simply:

“The blockchain is a decentralized server, a ledger of truth.”

He has already started publishing originals as Blockchain Video entries from a personal wallet/address, so future viewers can verify that a clip matches the creator’s signed source.

Practical steps: video validation + First Party Data

Trust affects conversion. With 24% of AI users already using AI shopping assistants, buyers (and their agents) will demand proof before they act. Brands can prepare by treating validation like a core asset, alongside First Party Data.

• Digitally sign high-value videos at creation time and store hashes for later checks.
• Publish a public verification page that explains how to confirm authenticity.
• Prioritize “source capture”: original files, timestamps, and chain-of-custody logs.
• Use Deepfake detection tools for monitoring, but rely on signatures for final proof.

5) Talent, Training, and the Interest-First Hiring Playbook

Marketing Teams need to stop confusing app fluency with craft

A common mistake in modern Marketing Teams is assuming that anyone aged 18–25 “gets social” because they grew up with the apps. Using platforms to watch content is not the same as building content that earns distribution. Many small businesses still hand social to a young relative because it feels logical. In practice, it often produces random posting, weak creative, and no learning loop.

“This is one of the hardest skills in the world.”

Training is the real moat (and it takes 1–2 years)

Social media marketing is difficult because it blends strategy, creative judgment, and fast testing. Even experienced brand marketers can struggle when the algorithm, formats, and culture shift. One operator points to a real training timeline: it can take 1–2 years before someone becomes effective, and the deepest expertise often comes from long tenure.

“The only people in my company that know it have been at the company for a decade.”

This is where Change Fitness matters: teams must build the habit of learning, shipping, and adjusting weekly, not yearly.

Interest-first hiring: test for intuition, not titles

As feeds moved from follower graphs to interest graphs, AI Visibility Works differently. Reach is earned by relevance signals, not by who already follows. Hiring should reflect that shift. The best candidates show curiosity about communities, comments, and context—why something works, not just what is trending.

• Hire for curiosity: asks “who is this for?” and “what would make them stop scrolling?”
• Hire for interest-graph intuition: can map affinity groups and content angles quickly.
• Evaluate output: prioritize views per piece and retention, not follower counts.

Practical evaluation: make candidates post and measure

Instead of guessing, teams can run a short, controlled trial. Candidates create content, publish, and iterate with AI-assisted testing. This mirrors how campaign optimization now works: rapid variations, fast feedback, and clear benchmarks.

1. Give a product and one audience segment.
2. Ask for 3–5 posts in native formats (video, carousel, text).
3. Track views, watch time, saves, and comments per post.
4. Require one iteration based on results.

Pod Based Models make this scalable: a small pod (strategy + creative + analytics) can learn faster, especially as agentic optimisation and purchase agents reshape tooling and hiring needs.

“First, you yourselves have to do it. You can’t judge something you don’t know.”

6) Six-Month Playbook: What to Do Tomorrow, Not Someday

Month 1: AI Visibility Works When Leaders Learn First

The next six months reward action, not opinions. Teams should block 50–100 hours for structured AI learning: tools, prompts, podcasts, and hands-on experiments. The point is not to “understand AI” in theory, but to build the reflex to ship, measure, and improve. As one operator puts it:

“First, you yourselves have to do it. You can’t judge something you don’t know.”

This learning time should feed a Predictive Strategy: what topics, formats, and hooks are likely to win next week based on what is winning today, not what a brand guide preferred last year.

Months 1–2: Publish Like the Algorithm Is a Marketplace

Distribution is being reset by Search Everywhere and Voice Search agents, which means discovery can happen in feeds, chats, and spoken queries—not just in one app. The simplest operating rule is volume with intent. The baseline experiment is clear:

“I just asked everybody to post twice a day on seven platforms.”

That target forces fast learning. AI can multiply output—some will even brag about “posting 4,000 times a day using AI”—but the playbook is to avoid noise and run tight Campaign Optimization loops. Every post should have one variable to test: opening line, thumbnail, length, or CTA.

Months 2–4: Measure Views per Asset, Not Vanity Metrics

Early on, the cleanest signal is views per picture/video. It helps decide what to boost, what to remake, and who to hire. The question to ask each week is:

“How many views does their picture and video get for you?”

In 2026, AI-assisted testing enables near real-time ad optimization, so creative that spikes organically can be turned into paid variants quickly, then iterated daily.

Months 3–5: Pod Based Models to Speed Learning

Pod Based Models outperform slow handoffs. Each pod should include strategy, creative, and analytics, with a small AI budget for testing. This structure compresses feedback, spreads insights across functions, and prevents “one team makes, another team guesses.”

Months 4–6: Provenance, Guardrails, and the Next Device Shift

As AI slop rises, brands should add authenticity guardrails and clear opt-out paths for audiences. High-value videos should be timestamped or registered to a wallet/blockchain to preserve provenance. Finally, the team should plan for the 6–7 year shift toward glasses: build transferable creative skills now, because the primary device will change and the game will reset again.

TL;DR: AI will remake marketing: it scales creative cheaply, elevates content over followers, threatens video trust via deepfakes, and rewards those who learn fast and post more.