Neural Networks: Capabilities, Challenges,and Threats

1.

Neural Networks:
Capabilities,
Challenges,and
Threats
Prepared by V. Rudyi
Group: NAMI 1-22

2.

Aim
To find out if neural networks are really
so dangerous for employment of people
in the sphere of art and text.

3.

Tasks
- To consider neural
networks and its types;
- To introduce an
imaging neural
network - Midjourney
and a text neural
network - ChatGPT.

4.

Neural networks
Neural networks are a type of
artificial intelligence modeled
after the structure and
function of the human brain.
They are designed to study
and recognize patterns in
data, and to make predictions
or make decisions based on
that data.

5.

Types of Neural
Networks
Artificial narrow intelligence is
built to perform a single task. For
example: Chess bot, Google
assistant.
Artificial general intelligence can
perform various types of tasks,
and find different ways to solve
them. For example: Chat-GPT,
Midjourney.
Artificial superintelligence, which
is more capable than a human,
does not yet exist, but it is under
development.

6.

Advantages of Neural Networks
Automation of time consuming
and labor-intensive tasks for
humans (image and speech
recognition, natural language
processing, and predictive
modeling).
Decision making by providing
information about large and
complex datasets (finance,
healthcare, and marketing).
Personalization of products
and services for individual
users based on their
preferences (movies, music
and products).

7.

1.An Imaging
Neural Network Midjourney.
It is a technology that is used
to generate, modify, and
enhance images. It uses deep
learning algorithms to create
photographs, illustrations,
pictures, and others.

8.

9.

10.

11.

Advantages
Disadvantages
The speed of query execution;
Errors in the construction of the
work;
Accuracy;
Free use.
Inability to create absolutely the
same character;
Inability to edit individual
elements;
Absence of completely new
ideas;
Inability to think over the whole
project carefully .

12.

2. A text Neural
Network -ChatGPT.
It is a type of neural network
to generate text data such as
automatic translation,
captioning, and text response
generation. They are based on
natural language processing
and machine learning models,
allowing them to learn and
improve with each new
challenge.

13.

14.

15.

16.

Advantages
Disadvantages
To perform tasks from
different areas;
If it cannot find information, it
creates it.
To reduce time searching for
information.
It does not understand the
context and particular details
of a project.
It cannot create completely
new ideas.

17.

Conclusion
Today neural networks should be considered as a tool and assistant to
perform various tasks and simplify life of professionals in the sphere of art
and text because neural networks have the ability to process huge
amounts of data quickly and efficiently, to do tasks very accurately, but
they are not able to understand the idea and particular details of a whole
project, and to create something completely new.

18.

References
https://www.techtarget.com/searchenterpriseai/definition/artificial-general-intelligence-AGI Artificial general intelligence (AGI);
https://www.simplilearn.com/tutorials/artificial-intelligence-tutorial/what-is-artificialintelligence - What is Artificial Intelligence: Types, History, and Future;
https://www.pcguide.com/apps/what-is-chat-gpt/ - What Is ChatGPT? – what is it used for?;
https://cloud7.news/ai/midjourney/ - What is Midjourney? Everything you need to know;
https://www.coursera.org/articles/types-of-ai - 4 Types of AI: Getting to Know Artificial
Intelligence;
https://dtf.ru/flood/1597231-pochemu-neyroseti-ne-smogut-zamenit-programmistov - Почему
нейросети не смогут заменить программистов;
https://trends.rbc.ru/trends/industry/63a192819a79478fae5762ad - Что такое ChatGPT и на
что он способен;
https://www.midjourney.com/ - Midjourney;
https://openai.com/blog/chatgpt - ChatGPT.
https://www.flaticon.com/ru/ - icons used in the presentation.

19.

Thank you for
your attention!