LivetechWestern cycleAnchors pending

Research-to-deployment tempo compression (publication → production)?

As of 2 Oct · 23:18 UTCResolves 31 Dec 20280 movers0 pressure moments
Timeline A · Timeline AReality
57%
Crowd 55%Gap 0

By end of 2028, AI research breakthroughs move from paper to working product in months, not years. A technique published on arXiv in January ships in a customer-facing system by June. Hugging Face rolls out new model architectures weeks after academic conferences. Frontier labs (OpenAI, Anthropic, DeepMind) publish a finding on Thursday and deploy it into their API by the following Monday. The feedback loop between academic insight and production code has collapsed into a single sprint cycle.

Timeline B · Timeline BReality
43%
Crowd 45%Gap 0

By end of 2028, the gap between published research and shipping product remains stubbornly long—12 to 24 months is still normal. Papers that appear at NeurIPS take years to see real-world use. Hugging Face support for novel architectures lags conference season by many quarters. Frontier labs publish research freely but keep deployment separate by weeks or months. Scaling headaches, safety validation, and engineering friction keep the research-to-product arc wide.

crowd 55
The quiet paththe one nobody’s watching

Regulatory freeze from a deployment-linked safety failure makes the acceleration question moot.

6%
The read

Where reality is heading

A, 80% bandCrowd today
Projected reality support to 31 Dec 2028A rises from 41% to about 52% with an 80% band of 33 to 71. B falls from 53% to about 44%. Quiet path near 4 to 6%. Crowd today: A 55%.100%7550250Crowd today · A 55%A ~52%B ~44%P ~4%NowJan 2027Jan 202831 Dec 2028
Reality crosses toward A in early 2028. Today’s crowd number sits at the top of A’s band, possible but optimistic. View as table

The movers, ranked by leverage

What would change our mind

Signals toward A
    Signals toward B

      Evidence