AI & HR: Insights to Keep You Ahead
A way to keep you up to date.
Welcome to the 89 newly FullStack HR people who have joined us since last week! If you haven’t subscribed, join 2419 smart, curious HR folks by subscribing here:
Happy Easter!
Before we dive into today’s topic, I want to highlight that you should watch the FullStack HR Podcast. I’ve been interviewing companies who are, in one way or another, working with HR Tech and AI and will be releasing these episodes during the spring.
Subscribe on Apple Podcast here.
Or search for FullStack HR in your favorite podcast player.
Last week I tuned into Lars talking to Q and AJ Thomas ⚡️, and Q mentioned in between that “someone should gather all the stuff that’s happening in this space.”
I thought I already do that; since 2014, HR and AI is a subject I’ve been monitoring, and where I try to read and follow what’s happening in the spaces. But I only share bits and pieces publicly.
So last week, I changed that by publishing a summary on LinkedIn. People seemed to like it, so I did the same yesterday.
And before you start eating candy or high-five the easter bunny, I thought sharing the highlights from these two weeks with you all would be useful.
(And yes, the headline is indeed created by ChatGPT and is optimized for getting people to open the email. Did it work?)
So here we go!
(PS. If you want to follow these updates in the future - add me on LinkedIn!)
Italy is banning (!) ChatGPT due to privacy concers. If there’s one thing I’ve stressed when talking about or writing about ChatGPT and HR is that you always, always, always should think about privacy. Italy takes it one step further.
CNBC has a good overview of what Italy and other countries and regions are doing.
The Washington Post reports that tech firms have been laying off people whose job was to look at the negative sides of AI. At the same time, we see increasing investments and development in the field, and one can only speculate what this will mean for the tools.
Let's hope for the best.
Want your own personal assistant? Now possible with AI! The dream of having an integrated personal assistant that solves every problem is lifelong for many people, someone who takes care of whatever you might help with. There have been numerous attempts trying to solve this with tech. It might now be possible with AI, and Lindy seems very promising. Sign up to get Beta access here.
A great article from Josh Bersin about SeekOut, a recruiting and talent intelligence vendor that has launched a new product called SeekOut Assist, which utilizes Generative AI to supercharge recruiting.
SeekOut is among several vendors, including Beamery and Eightfold, launching chat-based interfaces using GPT4.
But as Josh points out, and as I’ve pointed out several times before, we’ve only seen the beginning of this. Give it 1-3 more months, and the HR Tech scene will be flooded with products utilizing these technologies.
And one report that gives a short background about why we will see the flood is from Stanford University. They have released a report titled “AI Index Report 2023,” and it’s jammed with insights and data. It highlights that we’ve entered an exponential growth curve in machine learning systems, that AI research is through the roof, and that AI models get 1000000x (!!) more training today than they did a decade ago.
AI incidents and controversies are also up på 26x compared to a decade ago (and this is only up to 2021 - before Dall-e, Midjourney, ChatGPT, etc).
In forums, posts, and articles, there’s lots of talk about setting policies and guidelines around using AI in the workplace. Very happy to see the discussions around this.
As mentioned here before, I’ve created a free playbook template for creating ethical guidelines in the workplace. Scroll down in this article to fetch the playbook template (or, even better, read it before using it). The template is, of course, free.
Feel free to reach out with ideas or suggestions on improving it!
Beamery claimed they’re launching the “first generative AI for HR.” Deeply embedded in their products, they are now looking to redefine the user experience for their customers. It looks fantastic, and we’ll probably see many similar solutions shortly.
ariglad launched their HR Ticketing Workflow, and I’m a bit surprised that we haven’t seen more examples of this yet, as I think this is one of the most solid use cases - answering employees’ questions.
There’s an open letter circulating pledging to stop the development of AI beyond GPT-4. Behind the letter stands well-known AI scientists such as Max Tegmark and Stuart Rusell and tech people such as Elon Musk and Steve Wozniak. They make very fair points regarding the rapid development we’ve seen. “Powerful AI systems should be developed only once we are confident that their effects will be positive and their risks will be manageable.”
In forums and groups, I’ve seen a lot of discussion regarding using ChatGPT as a selection tool, e.g., drop your job description in the tool, then drop your interview notes and thoughts into the tool and let the tool decide if the candidate is a good fit or not. Don’t do that.
Goldman Sachs reported that up to 300 million jobs globally could be automated. The economists said that administrative workers and lawyers are expected to be most affected, compared to the “little effect” seen on physically demanding or outdoor occupations, such as construction and repair work. As I mentioned in a previous post this week, this roughly correlates with the report brought forward by The University of Pennsylvania and OpenAI.
Have anything else to share? Drop it in a comment below!