AI Innovation Lounge
AI Innovation Lounge
Prepare Your Organization for an AI-Powered Future
Prepare Your Organization for an AI-Powered Future
AI Research & Insights
Below are some of the studies and research I've bookmarked relating to AI. Each of these provides unique insights, findings, and more relating to AI. If you're looking for one I've referenced that is not on this list, please reach out and I'd be happy to share it with you.
Working with AI
Research on the impact of AI on the workforce & methods for working with AI
Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker...
Working paper to examine the performance implications of AI on realistic, complex, and knowledge-intensive tasks.
Author(s): Fabrizio Dell'Acqua, Saran Rajendran, Edward McFowland III, Lisa Krayer,Ethan Mollick, François Candelon,
Hila Lifshitz-Assaf, Karim R. Lakhani, Katherine C. Kellogg
Assigning AI: Seven Approaches for Students, with Prompts
This working paper explores seven approaches for utilizing AI in classrooms: AI-tutor, AI-coach, AI-mentor, AI-teammate, AI-tool, AI-simulator, and AI-student.
Author(s): Mollick, Ethan R. and Mollick, Lilach
A Roadmap to AI
This article introduces a 10 step roadmap to transformation with AI. From strategy to evaluation, it provides a comprehensive guide to follow.
Author(s): Stephen Treacy
Case Volkswagen Passenger Cars - Upskilling Strategy for Employees
An article in the book Artificial Intelligence Education in the Context of Work. This chapter focuses on a case study with Volkswagen Passenger Cars. (Unfortunately open source is not available)
Author(s): Judith Spirgi, Andreas Meier
TruthfulQA
This study proposes a benchmark to measure whether
a language model is truthful in generating answers to questions and tests different models.
Author(s): Stephanie Lin, Jacob Hilton, Owain Evans
GPTs are GPTs
This study explores the potential impact of large language models on the U.S. labor market. It revealsthat around 80% of the workforce could have at least 10% of their tasks affected.
Author(s): Tyna Eloundou, Sam Manning, Pamela Mishkin, Daniel Rock
Generative AI at Work
This study followed over 5,000 customer support agents at a Fortune 500 company for a year, using AI tools to aid their work. The outcome? An average productivity increase of 14%.
Author(s): Erik Brynjolfsson, Danielle Li, Lindsey R. Raymond
The Future of Jobs Report
This report explores how jobs and skills will evolve over the next five years.
Author(s): World Economic Forum
Impact of AI on the Workplace
An OECD paper analyzing case studies on the workplace impacts of AI. This found that 40% of companies reported skill requirement changes after introducing AI.
Author(s): OECD
ChatGPT & LLM Prompting Techniques
Research on various ChatGPT and large language model prompting techniques & methods
LLMs Understand and Can Be Enhanced by Emotional Stimuli
This paper highlights how different large language models like ChatGPT respond to emotional based prompts.
Author(s): Cheng Li, Jindong Wang, Yixuan Zhang, Kaijie Zhu, Wenxin Hou, Jianxun Lian, Fang Luo, Qiang Yang, Xing Xie
Unleashing Cognitive Synergy in LLMs : A Task-Solving Agent through Multi-Persona Self-Collaboration
Go beyond assigning a role. This paper presents a new method of prompt engineering that instructs the LLM to use multiple personas and have them collaborate
Author(s): Zhenhailong Wang, Shaoguang Mao, Wenshan Wu, Tao Ge, Furu Wei, Heng Ji
The Bold Promise Of Mega-Personas As A New Shake-Up For Prompt Engineering Generative AI Techniques
This Forbes article presents a new technique for prompting: Mega-Persona where ChatGPT can take on 100s or even 1000s of personas at once.
Author(s): Lance Eliot
From Sparse to Dense: GPT-4 Summarization w/Chain of Density Prompting
This paper highlights a new prompting technique to create detailed and entity-centric summaries.
Author(s): Griffin Adams, Alexander Fabbri, Faisal Ladhak, Eric Lehman, Noémie Elhadad
Prompt Reframing
This study looks at successful prompt instructions ad reframing techniques used to increase effectiveness and improve AI outputs.
Author(s): Swaroop Mishra, Daniel Khashabi, Chitta Baral, Yejin Choi, Hannaneh Hajishirzi
TruthfulQA
This study proposes a benchmark to measure whether
a language model is truthful in generating answers to questions and tests different models.
Author(s): Stephanie Lin, Jacob Hilton, Owain Evans
Zero-shot Temporal Relation Extraction
This study looked at ChatGPT’s ability on zero-shot temporal relation extraction. It looked at three different prompt techniques to evaluate outputs.
Author(s): Chenhan Yuan, Qianqian Xie, Sophia Ananiadou
Extractive Summarization
This paper highlights a thorough evaluation of ChatGPT’s performance on extractive summarization and compares it with traditional fine-tuning methods
Author(s): Haopeng Zhang, Xiao Liu, Jiawei Zhang
Large Language Model Workforce Impact
How will AI reshape the future of your job or workplace? As large language models (LLMs) like ChatGPT revolutionize different industries, many are curious about how our work landscape will evolve. A recent study revealed that 80% of workers may experience a shift in at least 10% of their tasks due to GPTs. Even more intriguing, 19% could see a staggering 50% of their tasks transformed.
3 Ways AI Impacts Job Skill Requirements
Did you know that 40% of companies reported skill requirement changes after introducing AI in the workplace? The OECD recently published a paper analyzing nearly 100 case studies on the workplace impacts of AI technologies. These are 3 key areas they identified where AI is transforming job skill requirements.
Critical Thinking Skills in an AI-Driven Future
Is AI making us lazy or too reliant on technology? Contrary to popular belief, AI actually seems to emphasize the need for critical thinking. A recent study, "GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models," found that critical thinking is among the least exposed skills to AI. AI also brings new critical thinking needs like evaluating accuracy and potential bias in AI-generated content.
AI Innovation Lounge | Copyright 2023