Humans are still better at summarising documents than artificial intelligence, an Australian government trial has concluded, despite the global hype around the technology, Crikey reported.
In a test conducted by Amazon for Australia’s corporate regulator, the Securities and Investments Commission (ASIC), it was found that humans outperformed the AI model used, the Australian news site story continued.
The test involved testing generative AI models before selecting one to look at and summarise five submissions from a parliamentary inquiry into audit and consultancy firms, the report went on. At the same time 10 ASIC staff were also given the same task, with judges overwhelmingly ruling that the human summaries beat their AI competitors on every criteria.
“This finding also supports the view that GenAI should be positioned as a tool to augment and not replace human tasks,” the report concluded.
- By Sean O’Meara
Also on AF:
Spotlight on Big Tech’s Power and Water Use Amid AI Surge
Asian Superpowers’ Rise in Critical Tech Research Ranking: ASPI
Study Suggests Ways to Overcome High Failure Rate in AI Projects
Chinese Tech Giants Pouring Cash Into AI Processors – FT
China, Tesla Pour Funds Into Assembly Line Robot Workers
