Keep in mind that this guide is like a snapshot. It shows what AI is now, how AI is used in workplace assessments, and what the implications for organizations are at one moment in time. The landscape is evolving so rapidly—sometimes hour by hour—that the technology is subject to sudden, significant change. Consequently, in this guide, we’ve emphasized ideas and strategy to help decision-makers navigate personality assessments in the era of AI.

What is artificial intelligence, or AI?

Artificial intelligence, or AI, refers to a computer system that imitates human thinking. Examples of tasks that require humanlike intelligence are perceiving, understanding language, synthesizing information, making inferences, solving problems, and making decisions. Making predictions is another way that an AI can mimic human thought processes. An AI that performs this task analyzes a lot of data and attempts to predict an outcome. It can refine its predictions over time or “learn” how to predict more accurately.

We should review a few essential terms related to artificial intelligence:

• Artificial intelligence, or AI – An artificial intelligence is a computer system that automates human thought processes.
• Algorithm – An algorithm is a step-by-step set of instructions or rules for a computer system to solve a problem or complete a task.
• Machine learning – Machine learning is a type of artificial intelligence in which computer systems learn from data and improve their performance without being explicitly programmed.
• Natural language processing – Natural language processing is a type of technology that allows computer systems to understand and use human language.
• Large language model – A large language model is a type of AI technology that uses natural language processing to produce content based on a vast amount of data. ChatGPT, for example, is powered by a large language model.

When many people think of AI, they probably imagine computers or robots that can speak and act like a human. Most AI systems today are computer applications. They are different from other types of programs or software because of how they complete tasks. Modern AI systems learn not by direct programming but by the experience of trial and error—one of the ways humans learn. In other words, machine learning is the attempt to use complex statistical modeling to allow the computer to learn from its errors.

Keep reading to learn more about the use of AI in talent management and, specifically, AI in personality tests.

Can AI predict personality?

Yes, AI can predict personality. Of course, that depends on what we mean by “personality.”

“If we think about personality as our core biology or our reputation, AI can predict that somewhat,” said Ryne Sherman, PhD, chief science officer at Hogan. “But not nearly as strongly as it can predict the kinds of things that we say about ourselves,” he added. AI can analyze various sources of data, such as text, speech, and social media activity, to calculate how someone might respond to questions on a personality assessment. So, to an extent, AI can predict the scores people are likely to get via personality assessment.

Targeted advertisements are a familiar analogy for the predictive ability of AI. If someone searches for camping gear and asks friends for advice about places to eat in Denver, it’s not a huge logical leap to assume they’re planning a camping trip to Colorado. An AI system might then show them ads for high-altitude tents or hiking shoes suitable for mountainous terrain.

In the same way, if an AI has personal data about someone, its machine learning algorithms can analyze that data to predict personality. Recent research showed that when an AI chatbot inferred personality scores based on the text of online interviews, it was overall reliable.² The easiest way to find out someone’s personality assessment scores, though, is to ask them to take a personality assessment!

“Technology drives many trends in our industry, some of which have more staying power than others,” said Allison Howell, MS, vice president of market innovation at Hogan. “The future of AI is incredibly exciting, but it’s important to remember that the technology is still in its infancy. As we explore potential applications, our commitment to quality and sound science remains a top priority.”

To be successful at prediction, any AI needs to learn from the right data, and it also needs feedback about whether it has made the right associations. If an AI makes a prediction based on incorrect data, the prediction won’t be accurate. That’s why traditional personality assessment should be just one of many factors that humans should consider when making any talent decisions.

How is artificial intelligence used in personality tests?

In personality psychology, artificial intelligence can be used to analyze responses to questions, identify patterns in data, and make predictions about personality characteristics. Whether it should do so raises questions about ethics and regulations, which we address later in this guide.

AI can use data either from personality assessments or from other sources, such as a person’s social media or web search history, to predict an outcome (for example, job performance). Some AI programs can even analyze audio and video to make inferences about an individual’s personality. However, when people make hiring decisions based on AI interviews or AI face scanning, bias is likely.³

One use for AI in personality tests is to help write questions or items for the assessment. Assessment companies could use AI to write questions or agree-disagree statements to identify how much conscientiousness someone is likely to show, for example. The accuracy of an AI’s outputs—in this example, assessment items or job performance predictions—depends on what data it uses for input and how many adjustments to its algorithms it has learned to make.

Do the Hogan personality assessments use artificial intelligence?

No, Hogan does not use AI in personality tests. “Our assessments are built based on traditional psychometric theories that have been rigorously researched and tested,” explained Weiwen Nie, PhD, research consultant. “This is why the way we build our assessments is the gold standard in the field of personality research.”

Ultimately, our goal in measuring personality is not only to provide insight about individuals but also to predict their workplace performance. Hogan has decades of scientific evidence showing how we achieve these goals.

If an organization claims to use AI in personality tests, but the AI doesn’t use evident algorithms or adhere to reliable psychometric theory, the results are not interpretable. Even if the results of the assessment describe personality characteristics, no one can know for sure if they are fair or even relevant if the algorithm used to generate them isn’t evident. This is what’s known as the black-box problem. When we don’t know what factors are driving an assessment’s predictions, the results are not useful for talent development—and they are unethical for use in talent acquisition. (More on that later.)

Now, Hogan does take advantage of some benefits of using AI in talent analytic processes. We use natural language processing, or NLP, to help classify job descriptions into job families. Natural language processing also helps us code subject-matter experts’ data when we perform job analyses. Each time, our subject-matter experts review the results and approve them. AI helps us automate these processes so we can create the best personality profile for a specific job. Using AI saves us time and resources and, in some cases, it even improves our analyses.

We believe that AI has the potential for more beneficial uses, which we are committed to exploring on an ongoing basis. Our assessments themselves, however, remain based on traditional psychometric theory.

Is it possible to “cheat” on personality tests using AI?

The answer is yes, but doing so is not advantageous. Our research shows that AI systems will usually answer personality assessment items with socially desirable response patterns—regardless of the context. For example, even if we prompt the AI to answer as if it were applying for a job as finance analyst or a salesperson, it will respond to the item in the same way.

The obvious responses make it easy to detect AI test results. In fact, Hogan has even built a tool that can determine if an assessment taker used ChatGPT to complete the Hogan personality assessments. We conducted a study to evaluate the tool’s efficacy at detecting cheating using 100 sets of assessment results simulating the response patterns of ChatGPT. To ensure that the tool would not falsely flag genuine responses, we also tested the tool on assessment results collected from 512,084 respondents prior to the emergence of ChatGPT. The results? Hogan’s tool detected 100 percent of ChatGPT responses and flagged zero percent of genuine responses.

Aside from being easily detectable, asking a computer program with no personality for help with a personality assessment is misguided. This type of dishonest candidate behavior is likely to be detectable during other stages of the hiring process too.

How can AI be used to improve talent management processes?

The benefits of using artificial intelligence to improve talent management processes are many. The practical applications of AI include informing decision-making in areas such as recruiting, onboarding, performance management, learning and development, and succession planning. It can summarize text, keep records, compare data, and assist with research, organization, and writing rough drafts.

“The strong suit of AI is in analyzing a large amount of data efficiently and making predictions based on that analysis,” said Chase Winterberg, JD, PhD, director of the Hogan Research Institute. He mentioned that an AI might help manage a large volume of applicants by prioritizing candidates, allowing humans to do more meaningful work instead of tedious, repetitive tasks. Similarly, AI chatbots might handle routine HR inquiries, while redirecting nuanced questions to humans.⁴ (Keep in mind that there are risks when using data from AI in making talent decisions, but we’ll mention those a little later.)

In talent acquisition, AI can help determine which competencies are most relevant for a job description. It can also help identify which personality characteristics are most important for performance on that job.

In talent development, an AI program might analyze worker time usage and make personalized suggestions for increasing efficiency or streamlining processes. An AI chatbot can even act as an on-demand virtual coach, helping people improve their performance at work. It also could provide customized career recommendations for a given personality profile or offer a reasonable series of steps to reach certain career objectives.

What are the risks of using AI in talent acquisition and talent development?

The risks of using AI in talent acquisition include making decisions using AI-generated information that is potentially biased. AI-driven decisions might inadvertently reinforce existing biases or create new ones, leading to unfair treatment of certain groups of candidates. For instance, an AI might incorrectly assume that protected characteristics, education level, or previous work experience is necessary to perform well in a job—and exclude candidates that don’t match its assumptions.

“Effective utilization of AI in talent acquisition requires a deep understanding of the data being used,” said Alise Dabdoub, PhD, Hogan’s director of product innovation. “Advanced statistical methods alone cannot compensate for inadequate research design. It’s crucial to have a comprehensive grasp of the data to avoid potential risks and biases in decision-making.”

The risks of using AI in talent development are a lack of inclusivity and accessibility. If an organization were to use AI for coaching, for instance, the AI might suggest that a person who belongs to a historically marginalized group behave like someone belonging to a group with more historical privilege. Not only is that not the best route for them, but it also perpetuates systemic biases. AI systems have an algorithmic process that they use to perform tasks, but that process isn’t always visible. Without a way to verify the algorithms, we cannot know for sure how an AI system is using its data.

Using AI in people decisions is perceived negatively by many US workers. Seventy-one percent of US adults oppose employers’ use of AI for making a final hiring decision.⁵ Even for reviewing job applications, 41 percent oppose employers’ use of AI.⁵ “There’s a risk of misinformation, confusion, and difficulty in making informed decisions,” Dr. Winterberg said. Talent management professionals must be very selective when using AI as a decision-making aid.

How can talent management professionals mitigate bias and prevent adverse impact when using artificial intelligence?

To mitigate bias and prevent adverse impact when using artificial intelligence, talent professionals can focus on data quality and maintaining transparency.

Focusing on data quality can help mitigate bias and prevent adverse impact with AI systems. If the data are low-quality or insufficiently diverse, then AI systems will produce outcomes that are low-quality or potentially biased. “We want to only consider variables that are job relevant, or important for succeeding in the job,” Dr. Winterberg said.

One way to know if job-relevant data are high-quality is to test or audit the AI system’s outputs. Rigorous AI testing can identify opportunities for improving data to generate an improved result. “Basically, you always need to be auditing AI systems for potential bias,” Dr. Sherman said.

Maintaining transparency into the decision-making process using AI systems can also help mitigate bias and prevent adverse impact. The need for transparency in any talent management process isn’t new. “Transparency is the cornerstone for building trust and ensuring ethical practices in talent acquisition,” said Dr. Dabdoub. “It is imperative to provide clear evidence that any selection system is job relevant, predictive of performance, and fair.”

If data that are generated by an AI system aren’t transparent, HR leaders should be wary of using them to make decisions in talent management. Organizations should create internal processes for identifying bias and build diverse teams for AI development until the technology meets quality standards.⁶

What regulations exist around using AI to make talent decisions?

At the present time, policymakers around the globe are still debating the best way to regulate using artificial intelligence in talent management. It’s challenging to decide how much risk to allow without reducing the benefits that AI can provide. However, laws already exist that basically apply to any employment decision, whether it’s a human decision or not. Dr. Winterberg pointed out the bottom line: “It’s illegal to discriminate on protected classes.”

We’ve listed several notable regulations here, and many more are being developed. Keep in mind that some items in the following list are best practices, while some are legal requirements:

• Ethical guidelines from the American Psychological Association state that only qualified individuals should interpret psychological test results, meaning that AI should not be used to interpret assessments.⁷
• The Society for Industrial and Organizational Psychology (SIOP) has published best practice recommendations covering the development, validation, and use of all hiring practices, including AI. SIOP also released a statement specific to using AI-based assessments for employee selection.⁸
• The European Commission has provided three overarching principles for what makes AI systems trustworthy. Artificial intelligence should be lawful, ethical, and robust.⁹
• The Uniform Guidelines are US federal recommendations for complying with Title VII of the Civil Rights Act, which protects employees and applicants from employment discrimination. The guidelines apply to all employment decision tools, including AI.¹⁰
• New York City adopted new rules about required bias audits for automated employment decision tools, which include AI.¹¹

Because regulations vary by jurisdiction, organizations should consult with legal experts to ensure legal compliance.

What are some ethical guidelines for using AI to make talent decisions?

The lines between lawful and ethical don’t always overlap. “AI technology can be built for one purpose and be used for other,” Dr. Sherman pointed out. “We’re at a place with AI that’s very similar to when scientists started colliding atoms.”

What makes using AI for talent decisions potentially unethical is the unknown element. This is the aforementioned black-box problem. To recap, different types of AI systems use algorithms that are either evident or hidden. If the algorithms are evident, it is easy for humans to know how the AI made its prediction. If the algorithms are hidden (as if they were inside a black box), we cannot see the steps that the AI took to reach its conclusion. This means the results could be irrelevant or unfair.

Common themes among most AI-related ethical guidelines are job relevance and transparency. It’s important to make sure that data the AI uses is relevant to the job. “It needs to actually be related to performance without having negative outcomes for any group of people who could succeed in the job. That sums up the basic implications for humans,” said Dr. Winterberg. It’s also important for AI use to be transparent in documentation and data privacy policies.^12,13 At Hogan, even though our assessments don’t use AI, we provide transparency into our validity and reliability, our logic, and how we predict workplace performance. We can show evidence for anything we do.

“The work we do has a profound impact on people’s lives, which is something we cannot take lightly,” said Howell. “Our clients trust us because our science is best-in-class. AI can help us serve our clients better, but applications absolutely must be developed as ethically as possible.”

The ethical thing to do when using AI is to publicize when and how it affects people. “Ethical considerations in AI usage demand transparency in communicating the impact on individuals,” emphasized Dr. Dabdoub. “It is crucial to publicize when and how AI decisions affect people. Keeping those affected informed is a fundamental aspect of responsible AI deployment.”

How should talent professionals select an assessment?

Organizations need to bring in people who are familiar with AI technology and can understand the potential implications for employees and risks for the business. They should also be able to provide proof that how they are using AI is fair, especially when it comes to AI in personality tests or other tools for making talent decisions.

Unsure about how to evaluate your assessment options? You’re not alone—let us help.

Contributors

We thank our contributors, listed here in alphabetical order, for sharing their expertise.

Alise Dabdoub, PhD, is the director of product innovation at Hogan. At Hogan, she has created an automated process for conducting cross-language equivalency of assessments, built norms, and conducted norm-shift impact analyses. She has an interest in quantitative methods, specifically the assessment of test and item fairness, and critical statistical methodology. She received her PhD in IO psychology from the University of Oklahoma.

Allison Howell, MS, is the vice president of market innovation at Hogan, where she leads the marketing and product development teams. She is passionate about leveraging Hogan’s best-in-class research to solve real problems for clients. She holds a master’s degree in science communication from the School of Journalism and Mass Communication at the University of Wisconsin-Madison.

Weiwen Nie, PhD, is a research consultant on the product innovations team at Hogan. He leads the application of the natural language processing and machine learning models to automate talent analytics processes. In 2023, he was part of a team that won an elite machine learning competition at the Society for Industrial and Organizational Psychology’s annual conference. He holds a PhD in industrial-organizational psychology from Virginia Tech.

Ryne Sherman, PhD, is the chief science officer at Hogan. He is an expert on personality assessment and data analytics, including the use of artificial intelligence and machine learning with personality assessment. Dr. Sherman has written more than 50 scientific papers, and he is the cohost of the popular podcast The Science of Personality. He received his PhD in personality and social psychology from the University of California, Riverside.

Chase Winterberg, JD, PhD, is the director of the Hogan Research Institute. In this role, he coordinates and communicates research to refine the theoretical foundation and understanding of best practices for implementing Hogan’s solutions to help solve organizational problems. He holds a JD from the University of Tulsa College of Law and a PhD in industrial-organizational psychology from the University of Tulsa.

Contact us for more information

Get Certified

The post Navigating Personality Assessments in the Era of AI appeared first on Peter Berry Consultancy.

“AI is where the future is, and we are moving into the future. You can’t avoid AI,” Ted said.

Let’s dive into a comparison of the strengths and weaknesses of AI coaching and human coaching respectively, as well as our guest’s advice for coaches.

AI Transformation Is Inevitable

Artificial intelligence, machine learning, and deep learning will become part of the fabric of just about every industry, including consulting psychology and leadership coaching. AI will affect our futures—and the outlook is bright. Ted pointed out that effecting successful technological change will draw upon the socioemotional skills of leaders to provide change management and psychological safety. It will also draw upon the expertise of psychologists during a time when every single organisational function is likely to be affected by AI.

Before we start imagining evil AI overlords dictating every minute of our working lives, it’s important to note that humans produce and control the content upon which AI is based. Ted explained AI “creativity” using two terms: generative AI and discriminative AI. Generative AI takes existing content and produces variations, such as ChatGPT inventing a comic book superhero based on Wonder Woman. Discriminative AI is a more predictive tool, which might identify employees who may become high-potential leaders or candidates who merit a second interview.

In the case of discriminative AI, an industry concern is that, while psychologists must follow guidelines, laws, rules, and regulations, AI tools behave the way they’ve been programmed. They don’t have accountability the same way that people do. Instead, AI has guardrails programmed into it by human engineers and data scientists who can choose to limit its access to information and its influence within the organisation.

Reiterating that AI is answerable only to its programming, Ted observed, “On the one hand, AI won’t save us from ourselves. On the other hand, it’ll reflect the best of us if that’s how we set up its content, and we are in control of that.”

AI Coaching vs. Human Coaching

In the consulting and coaching realm, AI coaching can offer some benefits that humans cannot. Machines are excellent at providing unrelenting reliability and processing data. In terms of the inability to like or dislike, they lack bias. They can help to train and support leaders and teams with instruction or data analysis. “It could do a lot in terms of developing people—not because it likes people, but because its imperative is to make people better at what they do,” Ted said.

Other benefits are that AI is always awake and accessible. It can’t get tired or distracted. It can learn a lot about you and make recommendations based on the data you provide to it, including how to achieve career goals. The advice is personalized. Even if it can’t contextualize or react to your emotions, it can choose a different option based on your response.

Now, human psychologists, consultants, and coaches currently have and always will have certain advantages over AI. If AI has some tools, a human coach has a wealth of tools, including AI. Humans understand how to leverage those tools relative to client needs.

An AI coach is likely to advise more broadly than a human. A human’s emotions and life experiences will allow greater specificity. Put another way, some scenarios will be so specific that where the AI might rely on data about it, the human has lived it.

Learning to position AI effectively will be a process. “We’re right at the dawn of all this. We just don’t know how good it’s going to get,” said Ted.

The Human Connection

A one-on-one, human-to-human interactive coaching session isn’t possible with an AI-powered coach. Perhaps surprisingly, however, there are some positives to coaching without the human connection. First, AI is without emotion, so how it feels about you cannot affect how it behaves. Second, AI cannot decide to lie to you because it has no metacognition. It is also superior to humans in its capacity to process information as an analytical function and to present information as a pedagogical function.

On the other hand, there are drawbacks to losing that human connection. One drawback is that data are messy, and AI using messy data might discriminate or generate racist, sexist, or homophobic responses. AI-powered systems require constant vigilance to achieve positive outcomes, and this need for oversight can be a downside.

Another drawback to AI coaching is the very absence of emotion. People tend to humanize machines. An AI coach or assistant, however, simply cannot answer many of the questions we might ask it. It could rate the pros and cons of a decision, but it can’t always know what is right.

Data privacy is another negative association with using AI tools. If an AI system construed personality information from an interview, for example, then it would be essential for the human to know how those data might be used for selection. An AI that collects information about organisational citizenship across platforms over time is another potentially privacy-violating example.

The human connection remains essential. Assessment feedback from a coach or psychologist trained in dealing with people and organisations will necessarily be superior to that provided by an AI. An AI system only knows what’s in its database. It can’t notice a reaction from a person and couch information in a certain way to be supportive because it cannot care. “Humans still have the upper hand, especially psychologists, in leveraging their expertise relative to what an AI system can do in terms of working with people,” said Ted.

Advice for Coaches

Ted suggested how coaches and consultants should respond to new technologies with a mindset focused on those they support.

Challenge your thinking about how to partner with AI in your journey as a consultant, psychologist, or coach. Choose to learn about ways to apply AI. Understand the ethical implications of working with AI systems in assessment and leadership development . Imagine how you would function inside an organisation that valued human behaviour enough to understand it through an AI platform.

Add value to the individuals in organisations and societies that choose to build out their AI function. “I understand the reasons to be pessimistic, but I don’t share them,” Ted concluded. “We can avoid issues like not accounting for algorithmic bias, for not treating people with dignity and respect as they deal with an AI system, for not understanding the possible environmental consequences of having an AI-based system. Because we are accountable, we can do that, and that sets us apart.”

Listen to this conversation in full on episode 71 of The Science of Personality. Never miss an episode by following us anywhere you get podcasts.

Contact us for more information Get Certified

The post AI in Psychology appeared first on Peter Berry Consultancy.

To understand why, it is important to understand how AI systems work. GPTs do not have “understanding” of medical science or poetry or the Roman empire in the sense that humans can understand these topics. Instead, these systems have learned associations between words and pieces of text.ⁱ When GPTs see a piece of text that says, “Write me an essay on the fall of Rome,” they see the words “write,” “essay,” and “fall of Rome.” They interpret that input as a request to generate a logically coherent (and possibly factually accurate) series of words and phrases connected to the text “fall of Rome” in essay format. The system learned these associations through a long series of trial-and-error efforts using massive computing power on vast amounts of text (i.e., the internet).

Because AI systems can take virtually any input and produce some logically coherent and possibly accurate output, it should come as no surprise that they can also take scores from Hogan assessments as inputs and generate text-based outputs. Given Hogan Personality Inventory scores as inputs, ChatGPT will provide interpretive text for those scores. For example, I asked ChatGPT to provide an interpretation of a score of 5 on the HPI’s Ambition scale, and it said:

“Your score in the Ambition scale is low, which suggests that you may not be particularly driven to achieve power, status, or wealth. You may be content with your current position and not feel the need to constantly pursue advancement or recognition. This can be a positive trait as it may allow you to focus on more important things in life.”

This is a logically coherent and reasonable interpretation. However, the interpretation does not say much about potential for leadership, the degree to which the scorer has a sense of direction in life, or the degree to which the scorer is comfortable in front of a large audience—all of which are captured by Hogan’s Ambition scale. Although the interpretation may seem accurate and valid, it may not be driven by any connection to Hogan at all, but simply by how the word “ambition” is used broadly in language.

Keep in mind that Hogan interpretative reports and guidance are based on empirical relationships between our assessments and outcomes. Scores on our assessments mean what they predict, and our reports reflect those relationships. This not to say that GPT-based interpretations of Hogan scores will not be valid or accurate now or in the future. In fact, training GPT models to reflect Hogan nomenclature is possible, and we are working on such tools currently.

But we must caution against using general artificial intelligence systems to generate interpretations of Hogan reports. Our own testing indicates that, at least on some occasions, the AI systems generate interpretations that are grossly incorrect and completely erroneous. While we may change our stance on this in the future as AI systems improve and more test results come in, for the time being we strongly recommend that any Hogan report interpretations come directly from Hogan or a Hogan-certified practitioner.

This blog post was written by Hogan Chief Science Officer Ryne Sherman, PhD.

Note

1. Some might argue that human understanding of these things is also simply association between words and text; we are not so sure we are ready to make that equivalency yet.

Contact us for more information Get Certified

The post Using AI to Interpret Hogan Scores appeared first on Peter Berry Consultancy.

Recently on The Science of Personality Live, cohosts Ryne Sherman, PhD, chief science officer, and Blake Loepp, PR manager at Hogan Assessments, spoke with Michal Kosinski, PhD, associate professor in organisational behaviour at Stanford University, about the evolving technology of artificial intelligence.

Michal’s primary research focus is studying humans in a digital environment using cutting-edge computational methods, artificial intelligence, and big data. He was also behind the first press article warning against Cambridge Analytica, the privacy risks they exploited, and the efficiency of the methods they use.

Let’s look at how AI language models have evolved, what AI-assisted communication might become, how AI affects the future of personality assessment, and whether AI language models can be creative.

The Evolution of AI Language Models

Within the next few months (as of March 2023), AI language models will become exponentially more capable and ingenious. How does that explosive growth happen?

The approach to the development of AI language models started with chess. At first, software engineers and data scientists fed AI chess programs with archives of chess games played by humans. Then they equipped two AI programs with a virtual chessboard and instructions for how to play without any human intervention. “For the first few million games, those models were completely stupid,” Michal said, explaining that the rate of play was millions of games per second. “But soon, after a few hours, what emerged was this alien, superhuman software that could play chess at a level completely unachievable to human players.”

Software developers and artificial intelligence specialists used the same adaptive strategy to teach AI models how to craft language. Humans learn language through conversation, context, and correction. They make mistakes, learn, and make mistakes more rarely over time. “At some point they stop making mistakes and reach new levels of language. The same approach was used to train ChatGPT and similar models,” Michal said. The AI programs were given sentences with one word missing, failed millions of times to fill in the blank correctly, and then began to get it right. After a few million dollars of electricity and a few billion sentences, Michal quipped, the programs showed language mastery at an extraordinary level.

The AI revolution originated by teaching machines to solve problems using the same strategies that we use to teach humans: reinforcement and feedback. At first, the machines make obvious logical mistakes, but then they don’t. “The AI is responding to you as if as if it’s another person, which is the most incredible thing,” added Ryne. Because computers can exceed humans in logical ability, they are well suited to both playing chess and using language.

AI-Assisted Communication

“AI is a revolution comparable with the invention of written language,” Michal said. Manual writing gave humans the ability to communicate across time, sometimes thousands of years in the past. Knowing how to use a stylus, quill, or pencil was an essential method for communication before computers. Now, knowing how to use a keyboard is essential. Very shortly, the same fundamental change will happen with AI language models, Michal predicted.

“I think that GPT is potentially a new language for humanity to communicate at speed and convenience unheard of and impossible before,” Michal said.

An AI language model won’t just help humans write emails. It will craft the perfect message in the language that is most readily understood for the recipient. Here’s how.

Imagine that Michal wants to send Ryne an email. An AI language model knows and remembers all the events of each person’s life and has consumed every piece of digital communication each has produced. If Michal asked the AI to send a message to Ryne, he could make the request in very few words as if speaking to a good friend with intimate knowledge of him. But because the AI knew Ryne at that same level, it could “translate” Michal’s message into the perfect form for Ryne. The AI could use not only Ryne’s preferred language, such as English or Mandarin, but also a highly personalised form of that language unique to Ryne.

“In terms of the potential for translation, it knows the meaning of what you’re trying to say. It can translate that into a meaning that somebody else can understand in the way they understand,” Ryne said.

Another sense of AI-assisted communication is searching the internet. You wouldn’t ask the AI language model to find a website for you; you’d ask it the question you wanted to learn. It would search all websites and tailor its answer to any length or depth for your individual understanding of the world.

AI in Personality Assessment

Artificial intelligence is great at knowing and remembering what has been written, both words and data. For an AI language model to predict personality based on language, you’d need to first collect a lot of quality data. Michal pointed out that AI language models already understand language, of course, and can translate words into analysable numbers. “They already understand psychological concepts like personality,” he said. These models have read texts written by introverts and extroverts and could theoretically detect, based on a fragment of a text, whether a person is introverted or extroverted.

Ryne imagined whether personality assessments of the future would have questionnaires and self-reporting. “One of the big questions surrounding this topic is to what degree I’m a willing participant in this endeavor,” he said. The quality of publicly available information versus data gained from individuals intentionally taking a personality assessment will differ substantially. The AI-assisted analysis would likely be higher quality in the latter case. Voluntary participation would also address questions of ethics.

Using big data models to predict personality characteristics is not a new notion. It has positives: it can analyse millions of people in a minute, and it can match people with compatible work or suggest workplace training and development. It also has negatives: it can be used to invade privacy or manipulate people. “As with many other technologies, we focus on the risks of the technology itself, completely forgetting that the real risk is in the intentions of the users,” Michal responded.

Artificial Intelligence and Creativity

A new fronter in AI language models is innovation and creativity. Humanity has taken generations to refine speech and writing. Individual humans spend over a decade learning to speak and write. AI language models have mastered written communication in a few years at a high level that continues to increase.

Michal compared AI creativity to human creativity in that most of us learn and combine elements of what we know or have experienced in new, creative ways. Perceiving computers as nothing but glorified calculators is short-sighted thinking, he said. That computers can incorporate and build elements into new results makes them fundamentally creative too.

“Many other animals are also creative in their own ways that we do not always recognize because it’s just not our type of art. The same applies to computers,” Michal said. “They learn from us, they learn from each other, and they become extremely creative with what they are good at—and they’re increasingly good at anything we ask them to do.”

Note: When ChatGPT (March 23 version) was asked to provide a quote in fewer than 120 characters about how it learned language, this was its response: “Words woven, sounds spoken, meanings grasped. A symphony of curiosity, immersion, and connection. Language learned, world unlocked.”

Listen to this conversation in full, and find the whole library of episodes at The Science of Personality. Never miss a new episode by following us anywhere you get podcasts.

Contact us for more information Learn about PBC Coaching

The post The Future Is Here: AI, Personality, and the Impact appeared first on Peter Berry Consultancy.