Dashboard
Overview of all tracked predictions and signals (updated daily)
Prediction data from pipeline — activity log is sample data from Day 2
Avg Confidence
54%
+15pp net today
Predictions
55
6 moved up today
Signals Today
24
from 270 scraped
Day
2
consecutive forecasts
Confidence Trends
Last 6 data points 34%
By 2030, AI coding tool revenue growth will have plateaued below 15% YoY, resembling the RPA hype cycle
19%
Junior developer hiring at FAANG companies will be HIGHER in 2029 than in 2024, not lower
47%
The median salary premium for 'AI engineering' skills will fall below $10K by 2029 as AI fluency becomes table stakes
17%
Open-source AI models will LOSE market share in enterprise coding tools to proprietary models between 2026 and 2030 -3
31%
AI-generated code will have a LOWER average CVE density than human-written code by 2029
47%
Total global software developer headcount will be HIGHER in 2030 than in 2025
64%
By 2028, at least 3 major companies will have publicly rolled back autonomous AI coding agent deployments due to quality/cost issues
61%
China will produce the leading AI coding tool by market share outside the US by 2029 +3
69%
Natural language will NOT replace code as the primary software specification medium by 2032
41%
The 'vibe coding' movement will produce a mass wave of catastrophic production failures by 2028, triggering regulatory intervention
67%
Cursor, the company, will not exist as an independent entity by 2028 (acquired or defunct)
68%
By 2030, >30% of professional developers will primarily use AI tools from a NON-US company (DeepSeek, Mistral, etc.) +5
32%
The senior engineer shortage narrative will prove wrong: AI will make senior-level judgment MORE accessible, not scarcer, by 2030
59%
At least one major AI lab (OpenAI, Anthropic, Google DeepMind) will suffer a sustained >6 month plateau in coding benchmark performance before 2028
50%
The biggest winner from AI coding tools by 2030 will be non-tech companies (banks, manufacturers, retailers), not tech companies
45%
By end of 2027, GitHub Copilot will lose its #1 market share position in AI coding assistants to a competitor
51%
By 2029, >60% of new production code at Fortune 500 companies will be initially generated by AI and then reviewed by humans
75%
At least one AI coding tool will achieve >90% on SWE-bench Verified by end of 2026
87%
The AI coding tool market will see a major consolidation wave: 5+ funded startups (>$10M raised) will shut down or be acqui-hired by end of 2027
90%
By 2030, the role title 'prompt engineer' will appear in fewer than 1% of software job postings, having been absorbed into standard engineering roles
35%
India will surpass the US in absolute number of developers actively using AI coding tools by end of 2028
10%
The 'AI software engineer' job title will be among the top 5 fastest-growing tech roles in the US by 2028
78%
By 2030, the bottom quartile of software developer salaries (adjusted for inflation) will be LOWER than 2025 levels in the US
68%
Staff+ engineer compensation at top tech companies will increase by >20% in real terms between 2025 and 2030
58%
By 2029, freelance developer hourly rates on Upwork/Toptal for routine web development will drop below $25/hr globally (from ~$50+ in 2025)
77%
A major AI-generated code vulnerability will cause a breach affecting >10 million users by end of 2027
88%
By 2028, >50% of enterprise SAST/DAST tools will have AI-code-specific scanning rules as a standard feature
59%
Prompt injection attacks targeting AI coding agents will become a new OWASP Top 10 category by 2028
58%
Fully automated AI test generation will cover >80% of test cases for new features at early-adopter companies by 2028
13%
By 2029, AI-driven continuous deployment pipelines will autonomously ship >40% of production changes at cloud-native companies without human approval
68%
AI code review tools will catch >50% of bugs that currently escape human code review, as measured in controlled studies, by end of 2027
56%
Pair programming between humans will decline by >50% at companies that adopt AI coding assistants, by 2028
46%
By 2028, the average time from idea to deployed MVP will drop below 1 week for solo developers using AI tools, down from 4-8 weeks in 2024
89%
By 2029, the majority of developers will spend more time specifying requirements and reviewing outputs than writing code directly +2
43%
Computer science degree enrollment in the US will decline >15% from 2025 peak levels by 2029
95%
Systems programming skills (Rust, C++, kernel development) will command a HIGHER salary premium relative to web development by 2030 than in 2025 +2
32%
By 2028, bootcamp graduates using AI tools will perform comparably to CS degree holders in their first year of employment, closing the historical gap
30%
The combined valuation of pure-play AI coding tool companies will exceed $200B by end of 2027
80%
Microsoft will bundle AI coding features into Visual Studio / VS Code at no additional charge, putting pricing pressure on the entire market, by end of 2027
18%
By 2029, at least one non-tech Fortune 100 company (bank, insurer, retailer) will build and open-source an AI coding tool that gains >10K GitHub stars
76%
GPT-5 or its successor will achieve >95% on HumanEval and >80% on SWE-bench Verified within 6 months of release
26%
By 2028, AI will be able to autonomously maintain a medium-complexity open-source project (>10K LoC, >100 issues/year) with minimal human intervention
33%
Multimodal AI coding tools that accept screenshots, diagrams, and voice input will capture >25% of the AI coding tool market by 2029 +4
94%
The cost per token for frontier coding models will drop by >90% between 2025 and 2028 +2
73%
The EU will pass binding regulation requiring disclosure of AI-generated code in safety-critical software (aviation, medical devices, autonomous vehicles) by 2029
15%
The EU's AI coding regulations will cause >20% of European startups to relocate or incorporate engineering teams outside the EU by 2030
28%
By 2030, India will have a domestically-developed AI coding tool in the global top 10 by revenue
54%
US-China AI export controls will fragment the AI coding tool market into two incompatible ecosystems by 2029
58%
By end of 2026, at least 3 Y Combinator batch companies will ship products built entirely by AI coding agents with <100 lines of human-written code
10%
By 2028, 'AI debugging' specialists — engineers who primarily diagnose and fix AI-generated code — will be a recognized job category at >10 major tech companies
77%
At least one AI coding tool will achieve >80% on SWE-bench Verified by April 30, 2026
66%
Anthropic will release a new Claude model (Claude 4 or equivalent) before May 15, 2026
74%
GitHub will announce changes to Copilot pricing or tiers before June 1, 2026
46%
At least 3 new AI coding startups will announce >0M funding rounds in April 2026
65%
The next major arXiv paper on AI code generation will report >75% on HumanEval+ before April 20, 2026
Signal Distribution
AI Coding
8
33% Job Market
5
21% Tooling
4
17% Security
3
12% Paradigm
2
8% Skills
1
4% Regulation
1
4% Confidence Heatmap
34
AI Coding
19
Labor
47
Compensation
17
AI Coding
-3
31
Security
47
Labor
64
Automation
61
Geopolitics
+3
69
Workflow
41
Regulation
67
Market
68
Geopolitics
+5
32
Labor
59
AI Capability
50
Market
45
AI Coding
51
AI Coding
75
AI Coding
87
AI Coding
90
Labor
35
Labor
10
Labor
78
Compensation
68
Compensation
58
Compensation
77
Security
88
Security
59
Security
58
Automation
13
Automation
68
Automation
56
Workflow
46
Workflow
89
Workflow
+2
43
Skills
95
Skills
+2
32
Skills
30
Market
80
Market
18
Market
76
AI Capability
26
AI Capability
33
AI Capability
+4
94
AI Capability
+2
73
Regulation
15
Regulation
28
Geopolitics
54
Geopolitics
58
Market
10
Labor
77
AI Capability
66
AI Capability
74
Market
46
Market
65
AI Capability
Category Health
Activity Log
06:00 Pipeline started
06:01 555 items scraped from 17+ sources
06:02 24 signals filtered (Gemini 3.1 Pro)
06:03 Report synthesized (9,132 chars)
06:03 P8 moved 78% → 82% (+4)
06:03 P1 moved 68% → 74% (+6)
06:03 P2 moved 74% → 77% (+3)
06:03 P4 moved 77% → 80% (+3)
06:04 Committed to almanac repo
06:04 Published to site
06:04 Pipeline complete (1.8 min)
Source Health
Hacker News
116 arXiv
90 TechCrunch
20 Ars Technica
20 MIT Tech Review
10 The Verge
5 GitHub Trending
9 Stack Overflow
— Metaculus
—