Unverified50% confidenceBenchmarkExact time
A joint study by BCG and Harvard found that when using GPT-4 for management consulting tasks, the human-AI collaboration group outperformed the human-only group by approximately 40%, provided that the boundaries of AI capabilities were accurately identified
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/30/2026
First Seen
Valid until: 11/28/2026
Sources
Related Entities
Related Claims
UnverifiedEven GPT-4-level Agents often achieve end-to-end success rates below 60% on complex tasks involving more than 5 tool calls.67% similarUnverified在Spider基准上，早期单模型方案准确率约为70%，GPT-4等大模型将其提升到80%以上，多智能体协作策略的系统已能达到85%甚至更高的执行准确率66% similarVerifiedA 2023 Stanford University study showed that multimodal models such as GPT-4V, when parsing flowcharts with nested conditional logic, had accuracy approximately 15–25 percentage points lower than equivalent textual descriptions.66% similarUnverified自2023年GPT-4发布以来，AI在HumanEval等主流编程基准测试上的通过率从不足50%跃升至90%以上65% similarUnverifiedThe generational capability gap between models is narrowing, with multiple models approaching or even surpassing GPT-4's reasoning advantage on certain benchmarks.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/825517API
curl https://kongchang.com/api/v1/knowledge/claims/825517MCP
get_claim(id=825517)