Hidden interlocutor misidentification in practical Turing Tests

Based on insufficient evidence, and inadequate research, Floridi and his students report inaccuracies and draw false conclusions in their Minds and Machines evaluation, which this paper aims to clarify. Acting as invited judges, Floridi et al. participated in nine, of the ninety-six, Turing tests staged in the finals of the 18th Loebner Prize for Artificial Intelligence in October 2008. From the transcripts it appears that they used power over solidarity as an interrogation technique. As a result, they were fooled on several occasions into believing that a machine was a human and that a human was a machine. Worse still, they did not realise their mistake. This resulted in a combined correct identification rate of less than 56%. In their paper they assumed that they had made correct identifications when they in fact had been incorrect.

I and I: immunity to error through misidentification of the subject

Abstract I argue for the following claims: [1] all uses of I (the word ‘I’ or thought-element I) are absolutely immune to error through misidentification relative to I. [2] no genuine use of I can fail to refer. Nevertheless [3] I isn’t univocal: it doesn’t always refer to the same thing, or kind of thing, even in the thought or speech of a single person. This is so even though [4] I always refers to its user, the subject of experience who speaks or thinks, and although [5] if I’m thinking about something specifically as myself, I can’t fail to be thinking of myself, and although [6] a genuine understanding use of I always involves the subject thinking of itself as itself, whatever else it does or doesn’t involve, and although [7] if I take myself to be thinking about myself, then I am thinking about myself.

Human misidentification in Turing tests

This paper presents some important issues on misidentification of human interlocutors in text-based communication during practical Turing tests. The study here presents transcripts in which human judges succumbed to theconfederate effect, misidentifying hidden human foils for machines. An attempt is made to assess the reasons for this. The practical Turing tests in question were held on 23 June 2012 at Bletchley Park, England. A selection of actual full transcripts from the tests is shown and an analysis is given in each case. As a result of these tests, conclusions are drawn with regard to the sort of strategies which can perhaps lead to erroneous conclusions when one is involved as an interrogator. Such results also serve to indicate conversational directions to avoid for those machine designers who wish to create a conversational entity that performs well on the Turing test.

The taxonomic status of pest species of Coptotermes in Southeast Asia: Resolving the paradox in the pest status of the termites, Coptotermes gestroi, C-havilandi and C-travians (Isoptera : Rhinotermitidae)

The taxonomic status of Coptotermes gestroi (Wasmann), C. havilandi Holmgren, C. travians (Haviland) and C. borneensis Oshima (Isoptera: Rhinotermitidae) is revised. The apparent discrepancy between the reported importance of C. havitandi in countries to which it has been introduced and the region from which it originated is shown to be due to misidentification and taxonomic confusion between C. travians, C. havilandi and C. gestroi. Based on an examination of specimens from Southeast Asia, two species are recognized, namely C. gestroi and C. travians. Coptotermes havilandi, described from imagos, is shown to be the same species as C. gestro described earlier from the soldier caste, and is designated a junior synonym. Coptotermes gestroi occurs from Assam through Burma and Thailand to Malaysia and the Indonesian archipelago, and has been introduced into other geographic regions, including parts of North and South America and the Caribbean. It is frequently found damaging wood in buildings, and is often intercepted outside its range in cargo onboard ships and sailing vessels, which is a likely mechanism for its spread to new geographical areas. Coptotermes gestroi has been misidentified in much literature as C. travians. Conversely, C. travians has been misidentified in recent literature in Peninsular Malaysia as C. havilandi and was redescribed from Borneo as C. borneensis, which is here designated a junior synonym of C. travians. It has a known distribution from Peninsular Malaysia to Borneo, and has not been found infesting wood in buildings. It is envisaged that the resolution of this taxonomic problem will enable the deployment of common pest management strategies for C. gestro the primary pest species of Coptotermes originating from Southeast Asia.

Deception-detection and machine intelligence in practical Turing tests

Deception-detection is the crux of Turing’s experiment to examine machine thinking conveyed through a capacity to respond with sustained and satisfactory answers to unrestricted questions put by a human interrogator. However, in 60 years to the month since the publication of Computing Machinery and Intelligence little agreement exists for a canonical format for Turing’s textual game of imitation, deception and machine intelligence. This research raises from the trapped mine of philosophical claims, counter-claims and rebuttals Turing’s own distinct five minutes question-answer imitation game, which he envisioned practicalised in two different ways: a) A two-participant, interrogator-witness viva voce, b) A three-participant, comparison of a machine with a human both questioned simultaneously by a human interrogator. Using Loebner’s 18th Prize for Artificial Intelligence contest, and Colby et al.’s 1972 transcript analysis paradigm, this research practicalised Turing’s imitation game with over 400 human participants and 13 machines across three original experiments. Results show that, at the current state of technology, a deception rate of 8.33% was achieved by machines in 60 human-machine simultaneous comparison tests. Results also show more than 1 in 3 Reviewers succumbed to hidden interlocutor misidentification after reading transcripts from experiment 2. Deception-detection is essential to uncover the increasing number of malfeasant programmes, such as CyberLover, developed to steal identity and financially defraud users in chatrooms across the Internet. Practicalising Turing’s two tests can assist in understanding natural dialogue and mitigate the risk from cybercrime.

Human microbiota in health and disease

Each human body plays host to a microbial population which is both numerically vast (at around 1014 microbial cells) and phenomenally diverse (over 1,000 species). The majority of the microbial species in the gut have not been cultured but the application of culture-independent approaches for high throughput diversity and functionality analysis has allowed characterisation of the diverse microbial phylotypes present in health and disease. Studies in monozygotic twins, showing that these retain highly similar microbiota decades after birth and initial colonisation, are strongly indicative that diversity of the microbiome is host-specific and affected by the genotype. Microbial diversity in the human body is reflected in both richness and evenness. Diversity increases steeply from birth reaching its highest point in early adulthood, before declining in older age. However, in healthy subjects there appears to be a core of microbial phylotypes which remains relatively stable over time. Studies of individuals from diverse geopraphies suggest that clusters of intestinal bacterial groups tend to occur together, constituting ‘enterotypes’. So variation in intestinal microbiota is stratified rather than continuous and there may be a limited number of host/microbial states which respond differently to environmental influences. Exploration of enterotypes and functional groups may provide biomarkers for disease and insights into the potential for new treatments based on manipulation of the microbiome. In health, the microbiota interact with host defences and exist in harmonious homeostasis which can then be disturbed by invading organisms or when ‘carpet bombing’ by antibiotics occurs. In a portion of individuals with infections, the disease will resolve itself without the need for antibiotics and microbial homeostasis with the host’s defences is restored. The administration of probiotics (live microorganisms which when administered in adequate amounts confer a health benefit on the host) represents an artificial way to enhance or stimulate these natural processes. The study of innate mechanisms of antimicrobial defence on the skin, including the production of numerous antimicrobial peptides (AMPs), has shown an important role for skin commensal organisms. These organisms may produce AMPs, and also amplify the innate immune responses to pathogens by activating signalling pathways and processing host produced AMPs. Research continues into how to enhance and manipulate the role of commensal organisms on the skin. The challenges of skin infection (including diseases caused by multiply resistant organisms) and infestations remain considerable. The potential to re-colonise the skin to replace or reduce pathogens, and exploring the relationship between microbiota elsewhere and skin diseases are among a growing list of research targets. Lactobacillus species are among the best known ‘beneficial’ bacterial members of the human microbiota. Of the approximately 120 species known, about 15 are known to occur in the human vagina. These organisms have multiple properties, including the production of lactic acid, hydrogen peroxide and bacteriocins, which render the vagina inhospitable to potential pathogens. Depletion of the of the normal Lactobacillus population and overgrowth of vaginal anaerobes, accompanied by the loss of normal vaginal acidity can lead to bacterial vaginosis – the commonest cause of abnormal vaginal discharge in women. Some vaginal anaerobes are associated with the formation of vaginal biofilms which serve to act as a reservoir of organisms which persists after standard antibiotic therapy of bacterial vaginosis and may help to account for the characteristically high relapse rate in the condition. Administration of Lactobacillus species both vaginally and orally have shown beneficial effects in the treatment of bacterial vaginosis and such treatments have an excellent overall safety record. Candida albicans is a frequent coloniser of human skin and mucosal membranes, and is a normal part of the microbiota in the mouth, gut and vagina. Nevertheless Candida albicans is the most common fungal pathogen worldwide and is a leading cause of serious and often fatal nosocomial infections. What turns this organism from a commensal to a pathogen is a combination of increasing virulence in the organism and predisposing host factors that compromise immunity. There has been considerable research into the use of probiotic Lactobacillus spp. in vaginal candidiasis. Studies in reconstituted human epithelium and monolayer cell cultures have shown that L. rhamnosus GG can protect mucosa from damage caused by Candida albicans, and enhance the immune responses of mucosal surfaces. Such findings offer the promise that the use of such probiotic bacteria could provide new options for antifungal therapy. Studies of changes of the human intestinal microbiota in health and disease are complicated by its size and diversity. The Alimentary Pharmabiotic Centre in Cork (Republic of Ireland) has the mission to ‘mine microbes for mankind’ and its work illustrates the potential benefits of understanding the gut microbiota. Work undertaken at the centre includes: mapping changes in the microbiota with age; studies of the interaction between the microbiota and the gut; potential interactions between the gut microbiota and the central nervous system; the potential for probiotics to act as anti-infectives including through the production of bacteriocins; and the characterisation of interactions between gut microbiota and bile acids which have important roles as signalling molecules and in immunity. The important disease entity where the role of the gut microbiota appears to be central is the Irritable Bowel Syndrome (IBS). IBS patients show evidence of immune activation, impaired gut barrier function and abnormal gut microbiota. Studies with probiotics have shown that these organisms can exert anti-inflammatory effects in inflammatory bowel disease and may strengthen the gut barrier in IBS of the diarrhoea-predominant type. Formal randomised trials of probiotics in IBS show mixed results with limited benefit for some but not all. Studies confirm that administered probiotics can survive and temporarily colonise the gut. They can also stimulate the numbers of other lactic acid bacilli in the gut, and reduce the numbers of pathogens. However consuming live organisms is not the only way to influence gut microbiota. Dietary prebiotics are selectively fermented ingredients that can change the composition and/or activity of the gastrointestinal microbiota in beneficial ways. Dietary components that reach the colon, and are available to influence the microbiota include poorly digestible carbohydrates, such as non-starch polysaccharides, resistant starch, non-digestible oligosaccharides (NDOs) and polyphenols. Mixtures of probiotic and prebiotic ingredients that can selectively stimulate growth or activity of health promoting bacteria have been termed ‘synbiotics’. All of these approaches can influence gut microbial ecology, mainly to increase bifidobacteria and lactobacilli, but metagenomic approaches may reveal wider effects. Characterising how these changes produce physiological benefits may enable broader use of these tactics in health and disease in the future. The current status of probiotic products commercially available worldwide is less than ideal. Prevalent problems include misidentification of ingredient organisms and poor viability of probiotic microorganisms leading to inadequate shelf life. On occasions these problems mean that some commercially available products cannot be considered to meet the definition of a probiotic product. Given the potential benefits of manipulating the human microbiota for beneficial effects, there is a clear need for improved regulation of probiotics. The potential importance of the human microbiota cannot be overstated. ‘We feed our microbes, they talk to us and we benefit. We just have to understand and then exploit this.’ (Willem de Vos).