ConferenceDigest
SessionsSpeakers
Briefs

Stakeholder intelligence reports

Themes

Cross-cutting meta-themes

Hypotheses

Claims pressure-tested

Ask AI
Sign in
Ask AI
ConferenceDigest

AI-powered conference intelligence

hello@confdigest.ioBuilt with Claude AI · Supabase · HeroUI
  1. Home
  2. /Digest
  3. /Hypotheses
Evidence-Based Analysis

12 hypotheses, pressure-tested.

We took testable claims from the conference and checked them against the full transcript corpus. Most didn't survive.

Evidence scorecard
Supported9
Partially supported3
Insufficient evidence0
Contradicted0
H1

AI Agents Will Displace Apps as the Primary Interface

Partially supported

Within three to five years, the dominant human-computer interaction model will shift from app-centric touchscreen interfaces to AI agent orchestration, rendering the current paradigm of lock screens, home screens, and siloed apps obsolete — and companies that do not build agent-accessible interfaces...

Confidence: medium

H2

95% of AI Pilots Fail Due to Human Factors, Not Technology

Supported

The primary barrier to enterprise AI ROI is not model capability, data quality, or tooling — it is human and organizational factors including leadership disengagement, employee distrust, inadequate change management, and the failure to kill perpetual pilots and fund production deployments.

Confidence: high

H3

Emotional AI Is the Critical Underdeveloped Frontier

Supported

Current AI systems are trained almost entirely on verbal content, making them blind to the 93% of human communication that is nonverbal — and this EQ gap means AI will fail to serve humans in the highest-value, highest-stakes contexts (healthcare, companionship, therapy, education) until emotional i...

Confidence: medium

H4

Social Disconnection Is the Defining Public Health Crisis of the AI Era

Supported

The loneliness and social disconnection epidemic — already accounting for up to 871,000 premature deaths annually per OECD estimates — will deepen as AI automation erodes the purposeful interdependence that creates social bonds, making social health investment one of the highest-ROI interventions av...

Confidence: high

H5

The Degree Premium Has Collapsed — Higher Education's Core Claim Is a Zombie Idea

Supported

The economic rationale for a four-year college degree as a guaranteed path to economic mobility has been empirically disproven — with the wealth premium for bachelor's degree holders collapsing from 250% (1930s graduates) to 42% (1980s graduates) and to just 6% for Black Americans — yet higher educa...

Confidence: high

H6

AI Governance Is the New Competitive Moat — Most Organizations Are Building It Too Late

Supported

By 2027, organizations without enterprise-grade AI governance infrastructure — agent identity systems, audit logging, data quality investment, compliance frameworks — will fail to scale AI successfully, and the 85% of AI investment that successful organizations direct toward governance and integrati...

Confidence: medium

H7

Independent Artists Are Building Better Business Models Than Major Labels

Supported

The independent music model — direct fan ownership, physical sales strategy, catalog-as-Netflix, platform-agnostic distribution — produces better economic and relationship outcomes for artists than major label deals in the streaming era, and the house party / direct-to-fan infrastructure emerging fr...

Confidence: medium

H8

AI-Assisted Creativity Narrows Ideational Diversity — A Measurable Harm to Innovation

Supported

When individuals use AI assistance for creative tasks, the resulting output converges toward statistically similar ideas, undermining the ideational diversity that drives genuine innovation. This measurable effect — documented in college application essays, design outputs, and content generation — m...

Confidence: medium

H9

Psychedelic Medicine Is the Most Underfunded Crisis Response in American Healthcare

Partially supported

Ibogaine and other psychedelic medicines demonstrate treatment efficacy for addiction and PTSD that dramatically exceeds existing standard-of-care interventions — with ibogaine showing 85% brain normalization within 48 hours versus 18 months for abstinence — and the continued Schedule I classificati...

Confidence: medium

H10

AI Tools Used Without Pedagogical Design Harm Learning Outcomes at Scale

Supported

Broad, unsupervised AI use in educational settings — particularly through consumer chatbots not designed for learning — demonstrably harms student cognitive development, creativity, and retention, while narrow, educator-guided AI use in purpose-built learning contexts shows genuine benefits. The dis...

Confidence: high

H11

AI Power Concentration Poses Democratic Risks That Regulation Is Failing to Address

Supported

A small cluster of AI companies and their investors are accumulating unprecedented control over information infrastructure, political processes, and economic systems — and current regulatory frameworks, weakened by successful tech lobbying, are inadequate to address the structural power concentratio...

Confidence: medium

H12

Human Augmentation Will Create the First Objectively Superior Humans — With Heritable Advantages

Partially supported

The convergence of gene editing, brain-computer interfaces, AI-optimized physiology, and augmented reality is producing human capability advantages that are not merely assistive but objectively superior — and CRISPR-enabled heritable gene edits already being made to embryos mean these advantages may...

Confidence: low