Revolutionizing Agriculture in Artificial Intelligence Through

Exploring the Potential of Machine Learning and Chat GPT

Vijay S

, Aarthi M

and Ana Raj

Department of Agricultural Extension, MITCAT Trichy, Tamil Nadu, India

Department of Extension Education, Assam Agricultural University, Jorhat, Assam, India

ARS, ICAR-National Research Centre on Equines, Haryana, India

Keywords: Agriculture, Artificial Intelligence, Chat GPT, Machine Learning.

Abstract: Revolutionizing agriculture is providing rebirth to agriculture sectors through the virtual applications in farm

practices. AI is steadily emerging as a part of the agricultural industry’s technological evolution. AI is an

umbrella term whereas ML is a subfield, that uses algorithms trained on data to produce adaptable models

using the past experience to predict the future, which can perform a variety of complex tasks. Chat GPT is

used for crop management and optimization, where farmers can use Chat GPT to get real-time insights on

weather patterns, soil health and crop growth predictions. The research was conducted at College of

Agriculture, Assam Agricultural University, Jorhat with the post graduate and Doctoral scholars of the

academic year 2023-24. 50 students were selected randomly and the data were collected using a well-

structured and pre-tested interview schedule. 82 percentage of the respondents prefer AI in receiving any kind

of information. The respondents have been found in highly using AI weekly twice (38%). Half of the

respondents (50%) anticipate that AI will have a boom in the field of technology in two years. Every

respondent of the study felt the need for training among the students regarding the field of AI. 94 percentage

of the respondents were found to be interested in receiving AI training right away. 58 percentage of the

respondents contemplated that importance of AI among the farmers was very high. Among the given

challenges that were anticipated in using AI, connectivity was ranked first by 24 of the respondents out of 50.

Research on AI in agriculture is still not sufficient and need to be explored. Meanwhile, there is a need to set

boundaries for using AI and accessing its functions.

1 INTRODUCTION

In the current scenario of global updating, cultivation

and transition in the agriculture sector is taken up by

virtual things. Especially the term ‘Digital

Agriculture’ means to cover broad aspects of various

disciplines under farming. Revolution in agriculture

is indeed due to the shortage of labor and increased

capitals. Revolutionizing agriculture is providing

rebirth to agriculture sectors through the virtual

applications in farm practices. Main focus is given on

reducing the time period and predicting the

anticipated outcomes of the future.

Technological practices have been developed

gradually in our day-to-day life. Indians have become

active internet users with 759 million to reach 900

million in 2025 (IAMAI). Internet penetration rate in

India was boosted fourfold in the last decade. It has

changed the life style and food pattern of humans

significantly. Artificial Intelligence plays a notably

vital role in that. The present status of mobile gadgets

and its peripheral usage on the earth is booming with

a great degree. It provides abundant information and

application in agricultural practices. Mobiles act a-

key-drivers in bridging the digital divide in farm

activities.

1.1 Artificial Intelligence

The quest to recreate human-like intelligence within

a computational substrate has obtained traction in

recent years, mostly due to the emergence of deep

learning methods and convolutional networks

(Schmidhuber, 2015; Bengio et al., 2015; Goodfellow

et al., 2016). Artificial Intelligence is a simulation of

110

S, V., M, A. and Raj, A.

Revolutionizing Agriculture in Artiﬁcial Intelligence Through Exploring the Potential of Machine Learning and Chat GPT.

DOI: 10.5220/0012882700004519

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Emerging Innovations for Sustainable Agriculture (ICEISA 2024), pages 110-115

ISBN: 978-989-758-714-6

Figure 1: Types of Artificial Intelligence.

human intelligence by machines. It thinks, senses and

acts like humans. According to Arend Hintze (2018)

the AI functions based on the listed types.

AI is steadily emerging as a part of the agricultural

industry’s technological evolution. AI powered

solutions effectively enables farmers with quality and

quantity output and ensures faster go-to-markets for

crops. Virtual platforms like face book, you tube and

digital pages exist everywhere, but there still exists a

gap to address. India is the one of the most populated

countries with a greater number of internet users

across the globe. Naturally, digital tools like Artificial

Intelligence, Machine learning, Chat GPT, Internet of

Things, Augmented and Virtual reality will easily

penetrate the country. Artificial Intelligence saves

agriculture from the instant climate change,

population growth, employment issue and food

safety. Agri GPT is one such example. According to

Wipro, a globally renounced company, AI is an

automation platform. Its robots used in agriculture,

will eliminate 80% of chemicals and will result in less

expenditure, up to 90% of the total cost. AI sprayers

drastically reduce the number of chemicals applied

for the particular produce to improve the quality of

the produce with cost efficiency. AI is a steadily

emerging and clubbing partner in the agriculture

sector along with technological evolution. AI can be

used as a ladder for agricultural development. The

global challenge is to increase food production up to

50% within the year 2050. As a step for achieving

this, AI provide powered solution in a fast manner,

like go-to market for the farm produces (Revanth

2022). Chat bots are trained to grasp the data in

human conversations and record the data based on

those dialogues, to learn how to understand what they

discussed, and finally come up with appropriate

responses. This is suitable to humans for both their

queries and problems (Pandey 2018). Researchers

trained the computer models to identify the traits of

the people based on the social media posts (Wu,

Kosinski and Stillwell 2015). However, AI is similar

to chat bots, that could be used to automatically

construct personalized messages for many people

using data obtained from their browsing history,

emails and tweets (Brundage et al., 2018). Reactive

machine is a part of AI which remembers and learns

the past success and failures, in case of the similar

situations that can happen in the future.

1.2 Machine Learning

AI is an umbrella term whereas ML is a subfield, that

uses algorithms trained on data to produce adaptable

models using the past experience to predict the future,

which can perform a variety of complex tasks.

Elements in ML algorithm process are

Representation, Evaluation and Optimization for

finding appropriate model for designing programme.

Its performance should be like humans but to reduce

the time and save the power. Key drivers in ML are

predictive modelling, automation, scalability,

generalization and addictiveness. In simple terms, the

process is that it learns, predicts and improves for

analysing the future demands. ML models used in

agriculture sectors in general are regression,

clustering, Bayesian models and clustering for a

variety of reasons.

1.3 Chat GPT

Chat GPT is an artificial language model developed

by Open AI on November 2022 in San Fransisco. It

uses ML for generating human-like conversation,

suggestion and feedback. GPT (Generative Pre-

trained Transformer) is the brain behind this. GPT-

3.5 is a language model, which can draw data from

the

internet to generate natural language responses.

Self-Awareness Theory of Mind

Reactive machines Limited Memory

Types of AI

Revolutionizing Agriculture in Artiﬁcial Intelligence Through Exploring the Potential of Machine Learning and Chat GPT

111

Figure 2: Elements in Machine Learning.

GPT-4 understands the context better, read between

the lines and understand the nuances. It is better

suited to respond to prompt situations that require

more complex and deeper understanding.

Especiallyin agriculture, Chat GPT is used for crop

management and optimization, where farmers can use

Chat GPT to get real-time insights on weather

patterns, soil health and crop growth predictions

(Destika et.al., Indonesia, National research and

innovation agency, 2023). GPT 4 and its image

description is mainly useful in agriculture sector for

detecting the disease and pests. Scientist team from

University of California, Berkeley and University of

Southern California explored in combining the NLP-

Natural Language Processing with AI driven chat

bots trained in GPT-4 language model to design

timely advisories. With the help of this, farmers

would get instant information on crop disease and

pest infestations. They evaluated the effectiveness of

those chat bots in reducing the agricultural losses. It

showed evident results of an average decrease in loss

by 5.5% and maximum of 16%. Researchers from the

University of California, San Diego (UCSD)

developed GPT-4 application through conversational

interface to answer farmer queries on crop growth,

pest control and other topics. It proved to be more

effective, accurate, fast, reliable and cost effective

than the traditional methods. The various GPTs make

the farmers pro-active and mitigate their problems

instantly. Now experts are venturing to expand these

services to even other disciplines of agriculture as

well.

Figure 3: Uses of GPT in Agriculture.

Elements in ML

Optimization Evaluation Representation

GPT in

Agriculture

Weed

Detection

Selective

breeding

Disease

detection

Yield

prediction

Water

Mnagement

Soil

Management

Livestock

Mnagement

ICEISA 2024 - International Conference on ‘Emerging Innovations for Sustainable Agriculture: Leveraging the potential of Digital

Innovations by the Farmers, Agri-tech Startups and Agribusiness Enterprises in Agricu

112

2 MATERIALS AND METHODS

The following study was conducted at Assam

Agricultural University, Jorhat. The university is a

roof of many colleges across the state. The research

was conducted at College of Agriculture, Assam

Agricultural University, Jorhat with the post graduate

and Doctoral scholars of the academic year 2023-24.

Ex-post facto study design was used for this research.

From the total of 262 students 50 were selected

randomly and the data were collected using a well-

structured and pre-tested interview schedule. The

scale was developed for the study which is designed

in a way to study about AI among the respondents

holistically.

3 RESULTS AND DISCUSSION

Table 1: About Artificial Intelligence among the

respondents.

Statements Frequency

(50)

Percentage

(100%)

1. Gende

Female 16 32

Male 34 68

2. Prefer to receiving information through

Yes 41 82

o 9 18

3. Frequenc

of usin

Dail

16 32

Weekl

twice 19 38

Weekl

once 10 20

Fortni

5 10

4. Years en

ed in usin

1 Yea

29 58

2 Years 11 22

3 Years 6 12

>3 Years 4 8

5. Anticipated

ear of boom for AI

1 Yea

12 24

2 Years 25 50

3 Years 8 16

>3 Yea

5 10

6. Need for training among the students

related to AI

Yes 50 100

o 0 0

7. Interested in receiving training related to

Yes 47 94

o3 6

8. Importance of AI amon

Farmers

Ver

h29 58

h12 24

Less 6 12

Ver

less 3 6

Table 1 provides information on awareness,

knowledge, preference, usage and training needs of

the respondents regarding AI. It shows that 68

percentage of the respondents were male and 32

percentage of them were female. 82 percentage of the

respondents prefer AI in receiving any kind of

information while 18 percentage of them were

skeptical in their preference for AI. Higher number of

respondents were found to be using AI weekly twice

(38%), followed by daily users (32%). 20 percentage

of the respondents engage in using AI weekly once

followed by 10 percentage of the respondents who

use AI at fortnightly interval. The percentage of

respondents using AI for the past one year were found

to be high (50%), succeeded by those who have been

using it for at least two years (22%). Less percentage

of the people were found to be using AI for three

years (12%) and very minimum of them were using it

for more than three years (8%).

From the table 1, it is also known that half of the

respondents (50%), anticipate that AI will experience

a boom in the field of technology in two years,

followed by those who had anticipated one year

(24%). Every respondent of the study felt the need for

training among the students regarding the field of AI

and were accepting that learning the know-how of AI

will be useful for better study experience and

performance in their stream. 94 percentage of the

respondents were found to be interested in receiving

AI training right away, while the remaining 6 percent

were hesitant in receiving such trainings. 58

percentage of the respondents contemplated that

importance of AI among the farmers was very high,

followed by 24 percentage of them who felt high

importance. Only 6 percentage of the respondents had

the opinion that the importance of AI among the

farmers is very less.

Figure 4 shows the six anticipated challenges in

using AI as perceived by the respondents. These

challenges were ranked by the respondents based on

their opinions and past experiences. Each challenge

was given a rank between 1 and 6 by all the

respondents

i.e., each challenge will be having 50

Revolutionizing Agriculture in Artiﬁcial Intelligence Through Exploring the Potential of Machine Learning and Chat GPT

113

Figure 4: Anticipated challenges in Using Artificial Intelligence.

responses. Among them, connectivity was ranked

first by 24 of the respondents out of 50. This proves

that connectivity in using AI as an obvious challenge

among all the others. The second rank was largely

given to decoding complexity by 12 respondents,

succeeded by ethical concern which was ranked third

by 16 of them. The language of some sites is simple,

whereas some are complex, of which the users might

have felt difficult to understand and interpret. Ethical

concern is a serious issue these days regarding AI

since the technology can be used for any kind of

purpose. The data and other information the user

entered might become a threat to him or her one day

which will automatically crush the trust of AI among

the users. The same number of respondents (14)

ranked Misinterpretations and Miscalculations as

fourth and creating huge dependency as fifth. With AI

providing information about everything, there might

be a chance where it can pose a threat to other sources

of information, especially books, journals, radio,

newspapers etc. Finally, the sixth rank was given to

reliability of the source by 12 respondents. Since the

source of information is not mentioned in AI, the un

shown authenticity of the information can be viewed

as a challenge.

4 CONCLUSION

Introduction of AI has changed the picture of human

life on earth. The sixth sense of humans has created

something equivalent to its capacity. This has its own

merits and demerits. It embodies machine intelligence

capable of perceiving and comprehending the

surrounding environment, culminating in the

production of the most efficient and productive

outcomes. AI has reduced human drudgery far more

than expected and proved to be precise and worthy.

But at the same time people are losing their jobs

because of AI. AI is a two-sided sharp knife which

should be handled with care. The use of AI in

agriculture is humongous when it comes to aspects

like weed management, selective breeding, disease

detection, yield prediction and management of soil,

water and livestock. It also encourages spray

reduction which yields good quality produce as well

as fertile soil. Leveraging the immense benefits from

these virtual tools to farmer doorsteps will produce a

valuable revolution in agriculture. It is also known

from the conducted research that the use of AI among

the students is also quite inevitable and more

progressive as every respondent felt the need to

explore AI. Many of the students were curious about

knowing the technical know-how of AI. Many were

already using it and some had a minimum two years

of experience in using AI. They also anticipate AI to

have a boom in a year or two. It is thus useful for

everybody in general and also for agriculture in

specific. Students, research scholars, scientists,

officials, all kinds of stakeholders, farmers etc. are the

beneficiaries of AI technology. To cherish the true

potential of this virtual tool in farming sectors it is

mandatory to understand its benefits and limitation

for its real time application. Some bottlenecks of AI

including errors in decoding, misinterpretations and

miscalculations, ethical concerns etc. needs to be

explored for effective upscaling. This calls for the

need of experts in the field who knows to handle these

tools since the manpower is scarce for the time being.

Meanwhile, there is a need to set boundaries for using

AI and accessing its functions. Streamlining AI is

important, as recently the European Union and

America had passed a new law on AI. Research on AI

0 5 10 15 20 25 30

Connectivity (R1)

Decoding Complexity (R2)

Relaibility of the source (R6)

Creating Huge Dependency (R5)

Misinterpretations and Miscalculations (R4)

Ethical Concerns (R3)

Anticipated Challenges in Using AI

ICEISA 2024 - International Conference on ‘Emerging Innovations for Sustainable Agriculture: Leveraging the potential of Digital

Innovations by the Farmers, Agri-tech Startups and Agribusiness Enterprises in Agricu

114

in agriculture is still not sufficient and need to be

explored. Further studies have to be conducted on

enforcing policies and laws regarding the

development and use of AI in every field including

agriculture for streamlining its use and reduce its

harms. Thus, the Ever Energetic and Enthusiastic AI

is highly potential and it can further be potentialized

with the help of human minds which created it in the

first place.

5 AREAS OF FUTURE

RESEARCH

• Comprehensive assessment of Artificial

Intelligence in agriculture and its allied sectors.

• Training needs in knowing AI to overcome the

developmental lacunas.

• Formulation of an ethical framework for using

Artificial Intelligence among the citizens of

India.

• Integration of Artificial Intelligence and

Indigenous Technical Knowledge for better

infusion and development.

• Study on the use of AI among the farmers,

researchers and the government officials and

ways to harness its true potential.

REFERENCES

Biswas, S., 2023. Importance of chat GPT in Agriculture:

According to chat GPT. Available at SSRN 4405391.

Zheng, Y. Y., Zhu, T. H., & Wei, J. I. A., 2022. Does

Internet use promote the adoption of agricultural

technology? Evidence from 1 449 farm households in

14 Chinese provinces. Journal of Integrative

Agriculture, 21(1), 282-292.

Revanth, S., & Venkat, S., 2022. Artificial Intelligence

Based Integrated and Distributed System for Preventing

Covid-19 Spread Using Deep Learning. International

Journal of Next-Generation Computing, 13(3).

Goyal, P., Pandey, S., Jain, K., Goyal, P., Pandey, S., &

Jain, K., 2018. Developing a chatbot. Deep Learning

for Natural Language Processing: Creating Neural

Networks with Python, 169-229.

Youyou, W., Kosinski, M., & Stillwell, D., 2015.

Computer-based personality judgments are more

accurate than those made by humans. Proceedings of

the National Academy of Sciences, 112(4), 1036-1040.

Brundage, M., Avin, S., Clark, J., Toner, H., Eckersley, P.,

Garfinkel, B., ... & Amodei, D., 2018. The malicious

use of artificial intelligence: Forecasting, prevention,

and mitigation, arXiv preprint arXiv:1802.07228.

Cahyana, D., Sulaeman, Y., Barus, B., & Mulyanto, B.,

2023. Improving digital soil mapping in Bogor,

Indonesia using parent material information. Geoderma

Regional, 33, e00627.

Arend Hintze., Douglas Kirkpatrick., Christoph Adami.,

July 23–27, 2018. The structure of evolved

representations across different substrates for artificial

intelligence. Proceedings of the ALIFE 2018: The 2018

Conference on Artificial Life, Tokyo, Japan. (pp. 388-

395).

Schmidhuber, J., 2015. Deep learning in neural networks:

An overview. Neural networks, 61:85–117.

Bengio, Y., LeCun, Y., and Hinton, G., 2015. Deep

learning. Nature, 521:436–444.

Goodfellow, I., Bengio, Y., and Courville, A., 2016. Deep

Learning. MIT Press, Cambridge.

Revolutionizing Agriculture in Artiﬁcial Intelligence Through Exploring the Potential of Machine Learning and Chat GPT

115