Be Careful What You Wish for ...

Be Careful What You Wish for ...

I'm getting nervous with the proliferation of all of these AI-infused chatbots that will change life and work as we now know it. Some of them are wrong. Really wrong.

For example, I just asked Bing’s AI chatbot to describe the difference between Performance-based Hiring and traditional behavioral interviewing. The answer was not even close to being correct! Bard was slightly better but still wrong.

This is critical especially when you're using these tools to guide your company's hiring decisions. There's enough bias and flawed thinking to begin with and getting bad advice is a recipe for disaster.

Now the Bing answer was quite concerning since I thought it was built on the same platform as ChatGPT, but it turned out that when I asked ChatGPT the same question I got the right answer.?

In case you’re interested, here’s the right answer cut and pasted from ChatGPT.?

No alt text provided for this image

Performance-based Hiring differs from traditional behavioral interviewing techniques in a number of ways :

  • Focus on Performance: In Performance-based Hiring, the emphasis is on what the candidate needs to do to be successful in the role. The job description is based on Key Performance Objectives (KPOs) - clear outcomes that the person in the role should be able to achieve. Traditional behavioral interviews, on the other hand, focus more on the candidate's past actions, behaviors, and skills without directly tying them to specific outcomes required for job success. (Note: this is the big difference that the other AI chatbots didn't understand. )
  • Past Performance Predicts Future Performance: Performance-based Hiring operates on the premise that the best predictor of future performance is past performance under similar circumstances. During the performance-based interview, candidates are asked to provide detailed examples of comparable past accomplishments. Traditional behavioral interviews typically ask candidates to describe how they behaved in past situations, but these situations may not be directly comparable to the challenges in the new role.
  • Bridging the Gap: The performance-based interview is designed to bridge the gap between what's needed to be successful on the job (the performance objectives) and the candidate's past accomplishments. This is achieved through a series of problem-solving questions where the candidate is asked to describe how they would handle job-specific challenges. Traditional behavioral interviews don’t explicitly make this connection.
  • Two-way Process: Performance-based Hiring recognizes that top candidates are evaluating the company and role just as much as the company is evaluating them. The process therefore emphasizes not just assessment but also attraction, selling the role and company to the candidate based on what motivates them. Traditional behavioral interviews may not put as much emphasis on this two-way street.
  • Structured Interview Process: Performance-based Hiring uses a structured interview process, which is designed to minimize bias and ensure all candidates are assessed against the same performance criteria. While some forms of behavioral interviewing are also structured, others can be less formal and more prone to bias.

Overall, while there are similarities between Performance-based Hiring and traditional behavioral interviewing, the differences lie mainly in the emphasis on job-specific performance, problem-solving related to the role, and the recognition of hiring as a two-way process.

Moral: AT chatbots are sometimes right and sometimes wrong.

Just like people.

__________________________________________

Lou Adler?is the CEO and founder of?Performance-based Hiring Learning Systems ?– a consulting and training firm helping recruiters and hiring managers around the world hire more leaders. Lou is the author of the Amazon top-10 best-seller,?Hire With Your Head ?(John Wiley & Sons, 4th Edition, 2021),?The Essential Guide for Hiring & Getting Hired ?(Workbench Media, 2013) and LinkedIn Learning’s?Performance-based Hiring ?video training program (2016). His new course on LinkedIn Learning -?Integrating Performance-based Hiring in into the Performance Management ?(2023) - in now available. Make sure you check out his?Hire with Your Head ?Virtual Book Club ?for the latest hiring ideas and trends.

Rando Tkat?enko

Developing Shorz - a fully AI automated video editing tool for Windows.

1 年

The quality of the outputs wildly differs between different AI-s. Bing is one of the weakest of the bunch currently. The most accurate is GPT4, which is part of OpenAI and ChatGPT(Paid Version). GPT4 is what we use for Talentscreener.ai - copilot for hiring and recruitment. It assesses candidates 90% similar to a professional human hiring manager. In terms of bias, it's actually much less biased than any human when assessing skills and qualifications. We filter out any personal, or cultural info before passing it to AI for analysis and our extensive tests have shown no statistical evidence of a bias in screening candidates. It simply does not see the image or any information about the person that could induce bias.

回复

I'm a Kenya Defence force officer,,,I need a job in Canada,,,please assist

回复
Ann Chepkirui

Attended Mount Kenya University - MKU

1 年

I'll keep this in mind

回复
Evelina Bern

Bringing Aerospike to Northern Europe

1 年

I'm the biggest AI advocate, especially for making hiring more efficient. Our team is constantly having chats with people who are all about incorporating AI into their hiring processes. It's exciting to hear, but also worrying to understand that many of them want to use AI to decide who to hire straight from the interview, just like that. Thanks for sharing this as a great reminder that we can't just put all our trust in AI, even though it's a game-changing tool to have in our toolkit.

回复

要查看或添加评论,请登录

社区洞察

其他会员也浏览了