Every deployment that shaped the framework.

8000 agents upskilled; customer likelihood to recommend increased by 14%

Saved 360000 legal hours annually; bank processes 12000 credit agreements per year

30% of GenAI projects abandoned after POC by end of 2025

95% of companies report GenAI pilots falling short

Forrester TEI composite: 333% ROI, $12.02M NPV over 3 years, payback under 6 months (figures are aggregate across 6 customers, not NAF-specific)

30 million unnecessary miles saved; 72 million lbs CO2 eliminated

29% efficiency savings possible with full POC implementation

70% of documents correctly extracted and interpreted

60-second claim processing; 50% operational cost reduction

Encoded gender bias through indirect markers

White-associated names preferred 85% vs Black-associated names 9%; demonstrates that off-the-shelf LLMs carry resume-screening bias

Violated Age Discrimination in Employment Act; screened out 200+ applicants

14% productivity increase; 34% improvement for novice workers

23% of employed respondents used GenAI for work weekly; 9% daily

15.5% avg efficiency boost; some radiologists up to 40% gains

Failed to predict rapid price swings across local markets

45% of high-maturity orgs keep AI projects operational 3+ years

Business AI adoption surged to 78% in 2024 (up from 55% in 2023)

85% of customer service leaders will explore or pilot conversational GenAI in 2025

68% of suppliers closed deals via AI; 35-day payment terms achieved

More than $55 million saved through waste reduction

3 billion total interactions; 98% find information they need; averaging 58M interactions per month

2.3M conversations in first month (67%); equivalent to 700 FTE

Chatbot swore at customers; criticized own company

Tribunal ruled airline liable for chatbot's false promises

98% advisor team adoption; document access jumped 20% to 80%

2.8 hours/week saved per employee; 80%+ daily usage rate

More than 90% of Lumen employees have a Copilot license; up to 50% productivity improvement on some tasks (e.g. re-platforming)

550K of ~780K staff trained on generative AI (AI Refinery, SynOps); AI usage monitored as a promotion input

3500 test files migrated in 6 weeks; 97% automation rate

33% suggestion acceptance rate; 20% line acceptance rate; 90% of developers report time savings (median 20%); 72% developer satisfaction

Failed to meet accuracy levels; problems with different dialects

Only 22% advanced beyond POC; only 4% creating substantial value

Eliminated wait times for 20% of contact center requests in 7 weeks

78% of companies use AI; only 18% of use cases yielded measurable ROI

IT most mature function; <30% of experiments will be fully scaled

Controlled experiment measuring enterprise productivity impacts

Measurable productivity gains across enterprise functions

12.2% more tasks; 25.1% faster; 40% higher quality

GenAI surpassed all other AI solution types in deployment frequency

Workers who use AI tools report higher job satisfaction

Automatic meeting logs; draft emails; rolled out to 15000 advisors

Companies with AI-led processes systematically outperform industry peers

Quadrupled from 3300 to 11000 licenses; scaling to all 120000 employees