1 Philosophy of generative artificial intelligence (AI).
Theoretical limitations and possibilities,
practical benefits and threats of large language models (LLMs)
The main foundations of generative AI are mathematical transformers or
large language models (LLMs) such as GPT-4 (OpenAI), Palm 2
(Google) and Luminous (Aleph Alpha). In order to understand the
limits and possibilities of these models, one
should deal with the philosophical (axiological, epistemological,
methodological and ontological) limits and possibilities of
mathematics and language, because these provide
the basic tools of these models. Philosophy of science is
the specialized discipline that discusses questions related to ontological,
methodological, epistemic, and axiological aspects
This lecture explains some philosophical (ontological, methodological,
epistemic and axiological) aspects of generative AI. Arguments are given to
explain why such impressive results can be achieved in the
grammatical and stylistic improvement and
in the translation of texts. Other arguments shed light on the reasons why
there has been no progress in semantics. In other words,
why these language models are unable to capture the meaning of the texts.
Last but not least, the logical-analytical coherence of the
texts created by the language models should be briefly addressed.
2 Practical exercises with generative AI tools. Search for
information, grammatical and stylistic improvement and translation of texts
This practice lesson shows how you can write a grammatically correct text
within seconds, even if you have difficulties with grammar. Moreover, the
text can still be written in different styles (simple, business, academic or
technical). In addition, this text can be translated into up to 95 languages
just as quickly.
3 References
Philosophy of
generative artificial intelligence (AI).
Theoretical limitations and possibilities, practical benefits and threats of
large language models (LLMs)
The focus of my lectures, publications, and seminars on the subject of
the Internet is the sensible use of the Internet. Participants should learn
to conduct effective internet research and create or maintain their own
internet presence.
Currently, extensive information about this topic is only available here
in German: Internet/Generative AI.
|