Developments in Artificial Intelligence (AI) have affected the way we work with increasing speed. We've witnessed the progression from OpenAI releasing ChatGPT as a research preview in November 2022 to Generative AI recreating Drake’s vocals in a viral song by April 2023. All in only 6 months.

This rapid development in this field has raised significant philosophical questions around the ethics of AI and its implementation in our organizations. On one hand, organizations are figuring out ways AI can enhance their organizational performance. On the other, organizations must prepare for upcoming legislation and address questions of fairness that bear very real impact on people.

Being an HR Tech platform, Searchlight is committed to advocating for ethical AI and acting as responsible stewards in its development and implementation. So, we dove headfirst into these challenging questions:

What are 'ethics,' and how do they relate to Artificial Intelligence?
What are the differences between AI, Machine Learning, and Generative AI?
What considerations should People and TA leaders make when implementing AI into their organizations, and where should they mitigate risk?
How is regulation shaping the future of AI, and what policies should we keep an eye on?
What are the applications of AI in People and TA?

In collaboration with Elizabeth Shober from People Tech Partners and Matt Charney from HR.com, we hosted a discussion to address these questions. My primary takeaway? People and TA leadership must be the decisive voice of reason at the table, aiding organizations in adopting AI to enhance performance and champion fairness and ethics.

If you missed the webinar and would like to watch it on-demand, scroll to the bottom of the page.

Why should ethical AI matter to you?

The impact of AI on People and Talent Acquisition is substantial. It can revolutionize how organizations recruit, manage, and develop their teams. Yet, like any transformative technology, it's crucial to ensure its design and use uphold the principles of fairness, transparency, and privacy. Failure to adhere to these principles can lead to legal penalties, damage to brand reputation, and most importantly, it can adversely affect the very people we aim to uplift with these innovations and rapidly accelerate the unfair practices we’ve work so hard to amend.

For HR and TA leaders, this presents an important responsibility - and opportunity: to guide the ethical implementation of AI within our organizations. Ignorance is not bliss; a lack of understanding about AI's ethical dimensions poses a risk that could compromise our ability to make informed and ethical decisions. Therefore, staying informed about ethical AI is not just a matter of staying current; it's a strategic imperative in the rapidly evolving landscape.

‍

“This is the opportunity where the People and Talent functions can really guide the organization in a strategic level and be the voice of how we use AI in our organizations...

As we are all beginning to use AI...it brings up a lot of ways in which we’re going to have to navigate through organizations using ethics and fairness. When we think of the composition of E-Staff [Executive Staff], the people on E-Staff that do this better than anyone is us. The Chief People Officer.” - Elizabeth Shober, People Tech Partners

3 Key Takeaways

‍

Adopt a Checklist for Ethical AI: People and TA leaders should adopt a guideline to evaluate ethical AI. Focus on these primary areas: Effectiveness & Safety, Fairness and Non-Discrimination, Transparency and Accountability, Privacy and Data Protection. Use this checklist as a critical decision-making tool when evaluating and selecting AI solutions.
1. Transparency and Accountability resonated the most with Elizabeth. In her role as a People Leader, she often hears requests for more transparency but explaining how an AI model reaches its conclusions can be tough. As leaders in People and Talent, it's crucial to communicate with candidates and employees how organizations are using AI to improve their work. It's important they know that we, as people, alongside technology, are driving the business forward. For Elizabeth, “it will be imperative to update this messaging to candidates and employees so that they feel as though nothing is happening behind their backs.”
2. For Matt, Privacy and Data Protection is the the most important component in the checklist. Currently, companies gather vast amounts of Personal Identifiable Information (PII) solely for targeted marketing. Then, as Matt shares, “now you factor in everything I’m doing at work and online into determining an algorithmic outcome...that scares me quite a lot” because it significantly magnifies behavioral targeting.
3. I rank Effectiveness & Safety and Fairness & Non-discrimination highest. But there's a trade-off between them. When we strive for equal success prediction rates across groups - or 'statistical parity' - the AI model's effectiveness often takes a hit. Also, because its difficult to explain an AI model’s conclusions, adding the need for 'explainability' may further compromise its effectiveness. In many ways, these two points in the checklist can be conflicting.
  
  I always see this checklist as an exercise of understanding what is the priority and what is most important? The one that I will always prioritize above the rest is Effectiveness. Does it actually work?

‍

[SHAREABLE] [PTP Webinar] Ethical AI (dragged).png

‍

Understand the Different Categories of AI: Not all AI is created equal. Understand the differences when assessing the implications of new and current tools for your organization. Remember, AI is not inherently neutral; it reflects the biases in the data it's trained on. Ensure risk is mitigated by investing early.
1. Rule-Based AI: The most basic form of AI - it operates by following pre-set rules. It's like a sophisticated flowchart: if this, then that. If you say 'Hello', it replies 'Hi there!'.
2. Machine Learning (ML): A subset of AI where systems learn from data. It identifies patterns and makes decisions based on that, without being explicitly programmed. Think of Netflix's recommendation system; it learns what you like by analyzing your past behavior.
3. Deep Learning (a subset of ML): This takes ML a step further; it uses artificial neural networks (inspired by our brain's structure) to learn from vast amounts of data. Imagine Netflix not just suggesting shows, but predicting the exact episode you'd like to watch next!
4. Generative AI: This is AI on a creativity spree. It can generate new, original content like music, images, or text. Continuing our Netflix example, Generative AI could potentially write a new episode or even a whole new show based on your viewing history and preferences.
Stay Informed about Regulations and Tech Developments: Regulations and tech advancements change rapidly. Staying current ensures your practices remain compliant and innovative. Following reliable publications like Charter and Product Hunt will keep you informed.

‍‍

The moment is now: People and Talent leaders need to be the champions of ethical AI

‍

“AI gives us the opportunity to really be leaders in our organizations. We have wanted to be at the table for a long time. We are the voice of fairness. We are the voice of ethics in our organizations, better than any other position in the company.

Let’s take it, let’s own it, let’s have the power around it. Let’s learn about it as it changes. Let’s go through it together.” - Elizabeth Shober, People Tech Partners

‍

AI will continue to reshape our organizations. However, the key to successful AI adoption is not just making our processes more efficient - it's about using technology to create a better world of work, one that values every potential and current employee, treating them with the fairness and respect they deserve. This is the right moment for People and TA leaders to be that critical voice at the table, ensuring fairness and ethics in AI implementation.

We understand that navigating the complexities of AI can be daunting. Searchlight is committed to partnering with you on this journey and empowering the leaders we work with, giving them the knowledge and tools needed for responsible implementation.

At Searchlight, we envision a world where hiring is a win-win for both employers and employees, with AI as a primary tool to realize this fairer world. From our company's inception, our commitment to ethical AI has guided our product development to ensure fairness in our outcomes. Searchlight’s AI exhibits no adverse impact on the people we serve. Clients like Verana Health and Zapier have trusted us to be reliable partners in their missions and have seen significant returns because of it. Our commitment extends beyond internal operations. We feel a responsibility as stewards of this technology, which led us to share our insights in this changing landscape.

‍

Thank you to People Tech Partners and HR.com for joining us in this timely discussion.

‍

About Searchlight

Our AI-driven Talent Assessment System helps our customers raise their talent bar objectively and cost-effectively. Unlike relying on subjective opinions to assess capabilities and culture, Searchlight’s ethical AI creates precise, personalized scorecards by learning from the top performers at your company. Customers like Coda, Coinbase, and Zapier have improved their Revenue Per Employee and Quality of Hire by simplifying their hiring process, lowering cost-per-hire, and reducing mis-hires. Book a demo to learn more.

Ethical AI in People and Talent

Why should ethical AI matter to you?

3 Key Takeaways

The moment is now: People and Talent leaders need to be the champions of ethical AI

About Searchlight

See how Searchlight can find the right person forevery role

Related posts

Our goal is to make hiring win-win for everyone

Ethical AI in People and Talent

Why should ethical AI matter to you?

3 Key Takeaways

The moment is now: People and Talent leaders need to be the champions of ethical AI

About Searchlight

See how Searchlight can find the right person forevery role

Related posts

AI in Hiring, Use Case #5: Objectively Measure Quality of Hire for the First Time—and Continually Improve It Through a Virtuous Feedback Loop

AI in Hiring, Use Case #4: Leverage Predictive AI for Individualized Onboarding

Introducing Searchlight Applicant Screening

Our goal is to make hiring win-win for everyone