A Safety & Health Chat with ChatGPT

By: Dave Johnson, Phylmar newsletter editor

Many of you are likely among the more than 30 million users of the global phenomenon ChatGPT (which gets roughly five million visits a day), an artificial intelligence chatbot released free to the public last November 30 by the small San Francisco AI company OpenAI. Try it out at https://chat.openai.com/chat. (That’s if you get can through. ChatGPT has frequent outages as it runs out of processing power.) ChatGPT acquires its knowledge by analyzing vast amounts of text posted to the Internet. It cannot distinguish between fact and fiction and does not source its information.

Simply type in any question and within seconds the chatbot rapidly fills the screen with the answer. That’s what we did recently in a quick “interview” with ChatGPT.

Why are some organizations not willing to prioritize workplace safety and health?

There could be various reasons why some organizations may not prioritize workplace safety and health. Some of these reasons may include:

1. Lack of awareness or education: The organization may not fully understand the importance of workplace safety and health or may not be aware of the risks and hazards in their particular industry.
2. Cost concerns: Implementing safety measures may require significant investments, which some organizations may not be willing or able to make, especially if they perceive safety measures as a cost rather than an investment.
3. Competing priorities: Organizations may have other priorities such as increasing productivity, meeting deadlines, or improving profits, which may take precedence over safety.
4. Complacency: If an organization has not had any serious accidents or incidents in the past, they may become complacent and believe that safety measures are unnecessary.
5. Lack of enforcement: Even if an organization has safety policies in place, they may not enforce them consistently or effectively, which can lead to a lack of accountability and a culture that does not prioritize safety.

Note the question is phrased as a fact – “why are some organizations not willing…” and the response is not fact-based but hedged with the use of the word “may” 10 times and the lead that begins, “There could be…”

ChatGPT is more authoritative when asked to describe a topic, a person, an event rather than speculate on “why.” There is no hedging with “what” and “how” answers:

What are the key features of a workplace culture that prioritizes safety?

A workplace culture that prioritizes safety typically has the following key features:

1. Leadership commitment: Leaders at all levels of the organization are committed to and actively promote a culture of safety and are willing to invest the necessary resources to create and maintain a safe working environment.
2. Employee involvement: Employees are actively engaged in safety initiatives and are encouraged to report safety hazards or concerns without fear of retribution. They are also given the necessary tools and training to work safely.
3. Continuous improvement: The organization has a system in place to identify and address safety hazards and risks and is continuously looking for ways to improve safety performance and prevent incidents from occurring.
4. Communication: There is open and transparent communication regarding safety issues, and employees are informed of changes to policies and procedures that affect their safety.
5. Accountability: The organization holds individuals and teams accountable for safety performance and takes action to address noncompliance or unsafe behaviors.
6. Positive reinforcement: The organization recognizes and rewards safe behaviors and performance and encourages employees to identify and report safety hazards or concerns.
7. Learning culture: The organization encourages a culture of learning and continuous improvement, and shares safety best practices and lessons learned to prevent similar incidents from happening in the future.

“Why” questions often require more complex answers that ChatGPT hedges on with the use of “may”:

Why are leading indicators not used more often by organizations to measure safety and health performance?

There are several reasons why leading indicators are not used more often by organizations to measure safety and health performance:

1. Difficulty in measuring: Leading indicators are often difficult to measure and quantify, especially for safety and health performance. They may require specialized equipment or expertise to track and may not be readily available or easily accessible.
2. Lack of understanding: There may be a lack of understanding among organizations about the benefits of using leading indicators, and how they can be used effectively to improve safety and health performance.
3. Emphasis on lagging indicators: Many organizations tend to focus on lagging indicators such as injury and illness rates, which provide a retrospective view of safety and health performance. While these measures are important, they do not provide a proactive view of safety and health performance.
4. Limited resources: Some organizations may not have the resources or expertise to develop and implement effective leading indicators. Developing and tracking leading indicators requires time, resources, and expertise, which may not be readily available in some organizations.
5. Resistance to change: Implementing leading indicators may require a change in organizational culture and processes. Some organizations may be resistant to change, and may prefer to stick with the status quo rather than adopting new practices.

Overall, while leading indicators have the potential to improve safety and health performance, their implementation requires a concerted effort from organizations to overcome these challenges.

This last sentence: “Overall, while leading indicators have the potential…” reflects conventional wisdom regarding the use of leading indicators. Conventional wisdom also runs through ChatGPT;s responses to our first two questions. It’s likely none of ChatGPT’s answers tell you much if anything you didn’t already know. That’s because the chatbot is using information from many articles, essays, blogs, Wikipedia, and other information it has picked up from Internet posts – and is common knowledge to subject matter experts.

Its information needs to be verified for accuracy. We find nothing inaccurate about its answers here. Predictable and standard fare, yes, but would you dispute the accuracy?

In its current form, ChatGPT can be viewed as a starting point, laying out basic information on any number of safety and health topics. (Imagine chatbot capabilities two years from now. Goggle, Microsoft, Meta, OpenAI and others are racing to develop and launch more powerful, sophisticated versions.) Topics appropriate to safety and health meetings, committee meetings, team meetings, pre-job huddles, prep work for management meetings can be researched starting with a chatbot such as ChatGPT, and followed-up, verified and complemented by your own research.

It’s a tool (now free) to add to your safety and health toolbox.