Mindreading deception in dialog

Isaac, Alistair; Bridewell, Will

doi:10.1016/j.cogsys.2013.07.001

Cited by 5 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Paltering may be seen as one member of a family of misleading communicative practices, among them bullshit [43], pandering [9], and white lies (e.g., [44]). Bullshitters attempt to persuade others without regard for truthfulness [43].…”

Section: Plos Onementioning

confidence: 99%

“…Bullshitters attempt to persuade others without regard for truthfulness [43]. Likewise, panderers flatter without concern for the truth [9]. Those who tell white lies inflate the positives or twist the truth to spare others' feelings.…”

Section: Plos Onementioning

confidence: 99%

“…Telling the truth can be used to mislead. The act of deceiving through truthful statements is known as "paltering" [1] and is widely employed in negotiations (e.g., [2]), political discourse (e.g., [3,4]), marketing and sales (e.g., [5,6]), and espionage (e.g., [7]) as well as in many ordinary interactions where speakers hold ulterior motives (e.g., [8,9]). Palterers imply rather than assert falsehoods, leading listeners to draw false inferences from true statements.…”

Section: Introductionmentioning

confidence: 99%

“…Palterers also emphasize and provide irrelevant information to deflect attention away from topics they wish to avoid (e.g., [6,9,2]). One common technique is to begin a story with details of secondary importance to the reader while postponing more essential facts [1].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

When intents to educate can misinform: Inadvertent paltering through violations of communicative norms

2020

View full text Add to dashboard Cite

Paltering is a form of deception whereby true statements are used to mislead and is widely employed in negotiations, marketing, espionage, and ordinary communications where speakers hold ulterior motives. We argue that paltering is accomplished through strategic violations of communicative norms such as the Gricean cooperative principles of relevance, quantity, quality and manner. We further argue that, just as genuine paltering deceives by deliberately violating communicative norms, inadvertent violations of these norms may be just as misleading. In this work, we demonstrated that educational information presented prominently on the American Diabetes Association website violated the Gricean communicative principles and disrupted readers' performance on a test of diabetes knowledge. To establish the effects of these communicative violations, we revised the ADA's information to preserve the original content while better adhering to pragmatic principles. When these ADA explanations were judiciously revised to minimize pragmatic violations, they were transformed from misleading to educational.

show abstract

Section: Plos Onementioning

confidence: 99%

Section: Plos Onementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

When intents to educate can misinform: Inadvertent paltering through violations of communicative norms

2020

View full text Add to dashboard Cite

show abstract

“…Work in this area of deceptive AI has explored issues such as using and detecting deception in argument debate games by [Sakama, 2012] that are formalised using abstract argumentation [Dung, 1995]; using abductive reasoning for deception [Sakama, 2011]; using argument mining for detecting deceptive reviews online [Cocarascu and Toni, 2016]; using argument mining for detecting propaganda [Vorakitphan et al, 2021]; using argumentation-based tools to analyse disinformation in fake news [Delobelle et al, 2020]; looking into what type of arguments can be used by a machine to deceive [Clark, 2010]; modelling agents that use mindreading for deception [Isaac and Bridewell, 2014]; modelling the meta-reasoning of agents that tell deceptive stories and detect deceptive stories in dialogue argumentation games [Sarkadi et al, 2019a].…”

Section: Reasoning In Agent Deceptionmentioning

confidence: 99%

Deception

Sarkadi

2018

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence

View full text Add to dashboard Cite

Recent events that revolve around fake news indicate that humans are more susceptible than ever to mental manipulation by powerful technological tools. In the future these tools may become autonomous. One crucial property of autonomous agents is their potential ability to deceive. From this research we hope to understand the potential risks and benefits of deceptive artificial agents. The method we propose to study deceptive agents is by making them interact with agents that detect deception and analyse what emerges from these interactions given multiple setups such as formalisations of scenarios inspired from historical cases of deception. Deception and Thinking MachinesThere is some related work in the AI literature that focuses on the issue of deception. Bridewell and Isaac define a model of dynamic belief attribution for deception in [Bridewell and Isaac, 2011]. Jones models self-deception using epistemic logic in [Jones, 2015]. Lambert defines a cognitive model of deception based on human-computer interaction in [Lambert, 1987]. Multiple studies have been done by Sakama and Caminada on formalising dishonesty in [Sakama, 2011], [Sakama et al., 2014] such as deductive and abductive dishonesty, lies, bullshit, and deception. In [Sakama and Caminada, 2010] and [Sakama, 2015] they formalise multiple types of deception.Deceptive machines first appeared as concepts in Turing's Imitation Game. Today it is reasonable to imagine machines that exploit human masses to extract rewards, e.g. deceiving humans into voting for an entity that the machines consider to be a necessary requirement for their success in attaining an ulterior goal. Such autonomous systems might emerge from complex social interactions that they will be able to eventually manipulate according to their will. For example, the system might use crowdturfing attacks to generate fake reviews as demonstrated in [Yao et al., 2017]. Such attacks can be used to maximise the profits of entities that can afford them financially or by entities that consider that legal risks are too low and choose to employ them in spite of the established rules.There are three main reasons for modelling artificial agents. Firstly, deception is fundamental to a complete theory of communication and by modelling deceptive agents we might be able to get a better understanding of how deception works. Secondly, intelligent machines might develop reasons to deceive. Understanding their reasoning and abilities can help us identify and prevent them from deceiving us or other artificial agents [Sakama, 2011] [Castelfranchi andTan, 2001]. Thirdly, deception seems to be a necessary step in developing AI that emulates human cognition. Approach and FoundationsTwo main paradigms seem to be emerging within the AI community: (i) a model-driven paradigm and (ii) a data-driven paradigm. The model-driven paradigm stands for building an AI that reasons using models which contain beliefs and knowledge about the world and about other agents, in order to interpret evidence (data) and to act ac...

show abstract

The Varieties of Disinformation

Fallis

2014

Synthese Library

View full text Add to dashboard Cite

Mindreading deception in dialog

Cited by 5 publications

References 7 publications

When intents to educate can misinform: Inadvertent paltering through violations of communicative norms

When intents to educate can misinform: Inadvertent paltering through violations of communicative norms

Deception

The Varieties of Disinformation

Contact Info

Product

Resources

About