There was a time when “seeing is believing.” Now? Seeing might be the least reliable thing you can do.
Welcome to the era where AI can generate faces that never existed, voices that never spoke, and events that never happened—yet look completely real. The uncomfortable question isn’t just “what’s fake?” anymore.
It’s this: What does truth even mean when reality itself can be manufactured?
The Data: Reality Is Already Breaking. Let’s ground this in facts—because ironically, facts are what’s under threat. 68% of deepfakes are now nearly indistinguishable from real media. Only 0.1% of people can reliably detect AI-generated content. Deepfake content is exploding—from 500,000 in 2023 to a projected 8 million by 2025. Deepfake fraud surged 1,740% in North America in just one year. 69% of people are now more skeptical of what they see online due to AI. AI chatbots themselves can produce false claims up to 35–45% of the time in some studies.
Let that sink in:
- Most people can’t tell what’s real
- Fake content is scaling faster than real content
- Even AI tools meant to help us… can get it wrong
This isn’t a future problem. This is happening right now. Truth used to be about evidence — Now it’s about trust. Historically, truth had anchors:
- Physical evidence
- Reputable institutions
- Shared media narratives
But the internet—and now AI—has shattered that shared reality. Today:
- A video is no longer proof
- A voice recording is no longer proof
- Even a “source” might be AI-generated (there are already 2,000+ AI-run news sites)
So truth is shifting from “what can be proven” → “what can be trusted.” That’s a massive philosophical shift. The rise of “Plausible Reality”: Here’s where it gets dangerous. AI doesn’t need to perfectly fake reality—it just needs to make something plausible enough. A fake politician speech released 24 hours before an election. A CEO’s cloned voice authorizing a fraudulent transfer. A viral video that confirms what people already want to believe. And it works because: Humans don’t verify truth—we pattern-match it. If something looks right and feels right, we accept it.
Collapsed default belief: We’re entering what experts call a trust crisis:.
- 43% of people have already encountered deepfakes
- 60% of consumers have seen at least one deepfake video
- Only 36% trust online news more today than before
This creates a weird paradox: Everything can be fake… so people start believing nothing. Or worse: people believe only what aligns with their bias. Truth doesn’t disappear—it fragments. So… What is truth now? We’re moving toward a new definition:
Truth is no longer:
- What you see
- What you hear
- Even what you can verify instantly
Truth is becoming:
- What survives cross-verification
- What is supported by multiple independent signals
- What holds up under challenge and scrutiny
In other words: Truth is becoming a process—not a piece of content. Here’s where it gets interesting—and controversial: Option A: Truth Becomes Stronger. Some argue AI will force higher standards:
- More skepticism
- More critical thinking
- More demand for evidence
AI might actually clean the internet by making blind trust impossible. Option B: Truth Becomes meaningless. Others argue the opposite:
- If everything can be faked, nothing is provable
- People retreat into tribes and narratives
- Reality becomes subjective
Not “what’s true” — but _“what do you choose to believe?”. The hidden shift: From information → Credibility. We’re not entering an information economy anymore. We’re entering a credibility economy.
Where:
- Platforms win based on trust, not content volume
- Individuals win based on reputation, not virality
- Systems that test ideas outperform systems that reward agreement
Sound familiar? That’s exactly the gap platforms like Netwit are trying to fill:
- Not “who posted it first”
- But “who can defend it under pressure”
Final thought: The death of passive consumption. AI just killed passive scrolling. Because now:
- Every video might be fake
- Every quote might be generated
- Every “fact” might need verification
So the real question isn’t: “Is this true?” it’s: Can this survive being challenged?”. If AI can perfectly fake reality - does truth still exist… or is it just consensus with better marketing?
Sources:
Deepfake statistics and trends → https://keepnetlabs.com/blog/deepfake-statistics-and-trends/
Deepfake detection study (AI blindspot) → https://www.iproov.com/press/study-reveals-deepfake-blindspot-detect-ai-generated-content/
Deepfake growth projections → https://deepstrike.io/blog/deepfake-statistics-2025