11 March 2024 in IEEE Access
AI programs, built using large language models, make it possible to automatically create phishing emails based on a few data points about a user. The V-Triad is a set of rules for manually designing phishing emails to exploit our cognitive heuristics and biases. In this study, we compare the performance of phishing emails created automatically by GPT-4 and manually using the V-Triad. We also combine GPT-4 with the V-Triad to assess their combined potential. A fourth group, exposed to generic phishing emails, was our control group. We sent emails to 112 participants recruited for the study. The control group emails received a click-through rate between 19-28%, the GPT-generated emails 30-44%, emails generated by the V-Triad 69-79%, and emails generated by GPT and the V-Triad 43-81%. Each participant was asked to explain why they pressed or did not press a link in the email. These answers often contradict each other, highlighting the importance of personal differences. Next, we used four popular large language models (GPT, Claude, PaLM, and LLaMA) to detect the intention of phishing emails and compare the results to human detection. The language models demonstrated a strong ability to detect malicious intent, even in non-obvious phishing emails. They sometimes surpassed human detection, although often being slightly less accurate than humans. Finally, we make an analysis of the economic aspects of AI-enabled phishing attacks, showing how large language models increase the incentives of phishing and spear phishing by reducing their costs.
22 September 2023 in PsyArXiv
Large language models (LLMs) have recently made vast advances in both generating and analyzing textual data. Technical reports often compare LLMs’ outputs with “human” performance on various tests. Here, we ask, “Which humans?” Much of the existing literature largely ignores the fact that humans are a cultural species with substantial psychological diversity around the globe that is not fully captured by the textual data on which current LLMs have been trained. We show that LLMs’ responses to psychological measures are an outlier compared with large-scale cross-cultural data, and that their performance on cognitive psychological tasks most resembles that of people from Western, Educated, Industrialized, Rich, and Democratic (WEIRD) societies but declines rapidly as we move away from these populations (r = -.70). Ignoring cross-cultural diversity in both human and machine psychology raises numerous scientific and ethical issues. We close by discussing ways to mitigate the WEIRD bias in future generations of generative language models.
9 January 2024 in Behavior Research Methods
We test whether large language models (LLMs) can be used to simulate human participants in social-science studies. To do this, we ran replications of 14 studies from the Many Labs 2 replication project with OpenAI’s text-davinci-003 model, colloquially known as GPT-3.5. Based on our pre-registered analyses, we find that among the eight studies we could analyse, our GPT sample replicated 37.5% of the original results and 37.5% of the Many Labs 2 results. However, we were unable to analyse the remaining six studies due to an unexpected phenomenon we call the “correct answer” effect. Different runs of GPT-3.5 answered nuanced questions probing political orientation, economic preference, judgement, and moral philosophy with zero or near-zero variation in responses: with the supposedly “correct answer.” In one exploratory follow-up study, we found that a “correct answer” was robust to changing the demographic details that precede the prompt. In another, we found that most but not all “correct answers” were robust to changing the order of answer choices. One of our most striking findings occurred in our replication of the Moral Foundations Theory survey results, where we found GPT-3.5 identifying as a political conservative in 99.6% of the cases, and as a liberal in 99.3% of the cases in the reverse-order condition. However, both self-reported ‘GPT conservatives’ and ‘GPT liberals’ showed right-leaning moral foundations. Our results cast doubts on the validity of using LLMs as a general replacement for human participants in the social sciences. Our results also raise concerns that a hypothetical AI-led future may be subject to a diminished diversity of thought.
8 November 2024 in Science Advances
Human forecasting accuracy improves through the “wisdom of the crowd” effect, in which aggregated predictions tend to outperform individual ones. Past research suggests that individual large language models (LLMs) tend to underperform compared to human crowd aggregates. We simulate a wisdom of the crowd effect with LLMs. Specifically, we use an ensemble of 12 LLMs to make probabilistic predictions about 31 binary questions, comparing them with those made by 925 human forecasters in a 3-month tournament. We show that the LLM crowd outperforms a no-information benchmark and is statistically indistinguishable from the human crowd. We also observe human-like biases, such as the acquiescence bias. In another study, we find that LLM predictions (of GPT-4 and Claude 2) improve when exposed to the median human prediction, increasing accuracy by 17 to 28%. However, simply averaging human and machine forecasts yields more accurate results. Our findings suggest that LLM predictions can rival the human crowd’s forecasting accuracy through simple aggregation.
8 February 2022 in Nature Communications
In repeated social interactions, individuals often employ reciprocal strategies to maintain cooperation. To explore the emergence of reciprocity, many theoretical models assume synchronized decision making. In each round, individuals decide simultaneously whether to cooperate or not. Yet many manifestations of reciprocity in nature are asynchronous. Individuals provide help at one time and receive help at another. Here, we explore such alternating games in which players take turns. We mathematically characterize all Nash equilibria among memory-one strategies. Moreover, we use evolutionary simulations to explore various model extensions, exploring the effect of discounted games, irregular alternation patterns, and higher memory. In all cases, we observe that mutual cooperation still evolves for a wide range of parameter values. However, compared to simultaneous games, alternating games require different strategies to maintain cooperation in noisy environments. Moreover, none of the respective strategies are evolutionarily stable.