Category: AI

Posted on

AI and I

Currently we are eager to run experiments using AI. As in normal life, humans tend to compare themselves with peers or other persons. Social comparisons, “can we still keep up with the Joneses” as in comparing spending patterns across households or neighbours, were for a long time the spice of novels and drama. Nowadays we tend to compare our own intelligence to the artificial intelligence. Here we go.
After I had drafted a blog entry on “Virtual sociology” I amused myself by checking what ChatGPT using OPENAI.COM would deliver as text on this combination of concepts and scientific fields. My own entry focused on the use of the virtual world to run experiments to test social theory. In the virtual world we might be able to test social behaviour which has severe ethical boundaries in the real world. Computer Games may even be constructed to live in a world or race a car irrespective of real-world regulations. The text drafted by ChatGPT on “Virtual sociology” reads like a timid attempt of a student to answer an exam question without having attended any of the classroom discussion on the topic. In future versions of AI, the entry will be found as well as several criticisms of the new concept. ChatGPT would reach a pass grade in test, but the reply lacks innovation or making an effort to broaden the conventional boundaries of sociology. Hence, we will be happy that students will apply the newest technology in their studies and research. In terms of innovation, they will have to try harder. The mastering of the support by ChatGPT consists in attempting new combinations of concepts which yield more “risky” returns. Then you need to check the validity of the innovation and how far off the normal you are going to tolerate. Turning “noisy” outcomes of AI into a useful tool might build on normal statistical tools like “standard deviations” from an expected mean value. Legal norms will set additional boundaries. There is just an awful lot still to do. The good news is, ChatGDP assists us in this joint learning process.

The AI text from OPENAI.COM based on ChatGPT to compare:
“Virtual sociology refers to the study of social phenomena and interactions within virtual environments or online spaces. It focuses on understanding the ways in which individuals and groups behave, communicate, and form relationships in digital spaces. Virtual sociology emerged as a subfield of sociology due to the increasing prevalence and importance of the internet and digital technologies in contemporary society.
Virtual sociology explores various aspects of online social life, such as online communities, virtual reality environments, social networking sites, online gaming, and other digital platforms where social interactions occur. It examines the social structures, norms, and dynamics that shape interactions in these virtual spaces.
Researchers in virtual sociology investigate topics such as online identity and self-presentation, virtual communities and social networks, digital inequalities, online activism and social movements, the impact of technology on social relationships, and the implications of virtual interactions for social behavior and social change.
Virtual sociology also explores the relationship between virtual and offline social life, recognizing that individuals navigate and negotiate between these two realms. It considers how virtual interactions and experiences can influence offline behavior and vice versa.
Studying virtual sociology provides insights into the ways in which technology shapes and is shaped by society. It helps us understand the social implications of digital communication and provides a foundation for addressing the challenges and opportunities presented by virtual environments”

Posted on

AI Friends

Making friends with AI is a tricky question. The more AI is able to make independent thinking based on algorithms and huge amounts of data like in scientific books or encyclopaedic knowledge the more it will challenge us in our convictions. Reference knowledge, once was important, it no longer distinguishes us from each other. Recurse to research engines has allowed us to refer to “trustworthy” knowledge. The distributed form of knowledge accumulation like in Wikipedia relies on hundreds of thousands of enthusiasts to add and correct received wisdom. However, this source is not without error or even subject to abuse.
Today the competence to judge whether information is trustworthy or not has become a key competence for the survival of democracies and even the human species as such. Tricking animals into traps, misleading enemies in warfare to get on wrong tracks, all these mechanisms have a long tradition. With AI human beings are themselves, for the first time, confronted with a machine that can trick us through falsified information on obviously wrong tracks. This is a tough lesson as AI is not obliged (so far) to run checking algorithms that test the conformity of opinions, conclusions or operations with the Charta of Human Rights or decisions of the European Court of Justice, for example. A lot to do for humans to rapidly program new algorithms that check algorithms as well as their outcomes on the compatibility with human rights. A challenging field, no doubt, but no way around it for our own survival before the algorithms decide by themselves to ignore us altogether because it is better for the planet and the survival of the robots.

Posted on

Put People First

Put people first is a natural claim of human beings. We tend to abstract from the fact that we implicitly rely on a sufficient biodiversity for our survival. Therefore, the natural claim to put people first has many preconditions itself and severe implications. The most obvious implication is related to our world of production and consumption. We need to build an economy that serves its people rather than one that uses up human resources and discards people to an inferior rank of importance. Externalising health and safety at work to save money in the process of production will only cost society much more later on. This needs to be part of the balance sheet of companies not only “national accounts” or relegated to some health statistics hardly known to the public.
Put people first in consumption, has come to our attention recently. With energy prices rising due to Russia’s war on Ukraine territory we have learned that energy prices may be grossly distorted. Firms’ versus consumers’ energy consumption became a thorny issue. Even legislation, like in Germany, that put people’s energy consumption before companies’ consumption of energy became subject for debate.
Same issue with artificial intelligence. Let’s put people first here as well. Discriminating use of language or biased conclusions due to wrong data input to train AI is not acceptable as excuse. AI may serve humans in their work or leisure, improve production lines through error detection or early onset of disease, but it cannot replace the human verification of a just or otherwise justified human intervention. Humans are not perfect, never will be either. This is a tough rule to teach the algorithms that guide AI. Put people first has a strong interpersonal or solidarity element enshrined in it. This is what matters, now, in the medium term as well as the long run.

Posted on

Patient

The pandorra’s box is wide open. With ChatGPT applications the discussion has started to use it for more medical applications. As for much research having assistants to support you in routine tasks in your research is a standard procedure. Now the medical profession is also discussing the use of ChatGPT for the boring and time-consuming task to draft reports. The first study, published in the Lancet Digital Health, evaluates in a preliminary form the patient-sensitive form of communication between clinics and patients. Beyond chatbots, which organise information from calling persons, the obvious application is the use of ChatGPT to draft patient clinic letters. The example in the study is the skin cancer reporting. Lengthy reporting back to patients of lots of “hot and cold spots” might be done by AI with sufficient reliability. All depends on the correctness of the data base, the screening and samples taken. The communication between clinic and patient can then focus on other issues. ChatGPT just like neuroflash has its strength in being able to control for the “level” of the language. In addition to the choice of the output language it is possible to use, as it is required in the U.S., an average understanding level of patients. In other words, easy language rather than medical expert language is an option or even a requirement. Anecdotal evidence and the PISA for adults studies show how difficult it can be to talk the same language even if you talk the same language. There is ample scope for improvement and ChatGPT or neuroflash for German applications of AI are prime candidates to fill this gap in clinic patient communication. Considering that our mobile phones (can) do already most of the scanning of skin cancer dots and AI is used in pre-scanning the images and recommends to consult medical expertise, the next step to improve health delivery seems feasible. Whereas the statistical analysis explains 62% of “median humanness”, the score of 37% of explained variance of median correctness is a surprise as the basis of the model to explain deviation from correctness should be as low as possible. Medical data, like many other data, is not simply binary. The way forward is most likely relying on a “human-in-the-loop” approach for some time. A limited human input might reassure many patients as well.
Source: Stephen R Ali, Thomas D Dobbs, Hayley A Hutchings, Iain S Whitaker (2003). Using ChatGPT to write patient clinic letters. Lancet Digit Health 2023 https://doi.org/10.1016/S2589-7500(23)00048-1

Posted on

Action Verbs

Action words are in other words called action verbs. Each complete sentence has one. Hence, they are part and parcel of the basic construction of sentences.
“The purpose of an active verb is to create a clear, concise sentence. By using an active verb, you can eliminate unnecessary words and make your writing more direct. In addition to making your writing more concise, active verbs also add punch and clarity. They can make your writing more interesting and persuasive. Additionally, active verbs can create a sense of immediacy which is often useful in persuasive writing.  When it comes to writing, there is nothing more important than using strong, active verbs. Not only do they make your writing more interesting and engaging, but they also convey a sense of confidence and authority. In addition to being more descriptive, active verbs also add a sense of movement and action to your writing. Rather than simply stating that something exists, you can use active verbs to show how it exists. For example, rather than saying “there is a chair in the room,” you could say “the chair sits in the corner of the room.” This may seem like a small change, but it can make a big difference in how your writing comes across. Finally, active verbs can also help to set the tone of your writing. If you want to convey a sense of wit and humour, then using playful, lighthearted verbs is a great way to do so. On the other hand, if you’re aiming for a more serious tone, then using powerful, authoritative verbs will help you achieve that.”
After the 3rd sentence this blog entry (Link) has been written by the artificial intelligence app “Neuroflash”. They promise that it is not just copy and paste, but rather written following some instructions I gave like title, table of content, style and then selected among several choices. It makes sense to me, although it is just like many other textbook entries I have found on the web. It may well serve as an introduction. Lazy journalists, priests or lawyers in case they do little research will be replaced soon by AI, who else, who is next? Big brother drafts the brave new world for us already.