The annihilation of humanity by artificial intelligence? Fortunately, this scenario still belongs to the realm of science fiction, even though we are currently the closest to it historically. Whether we like it or not, artificial intelligence is here and it is not going away - on the contrary, it is becoming more and more advanced, and to stubbornly ignore it in certain sectors would be tantamount to trying to fight windmills. Let's take a look at how today's AI can help you in programming, cybersecurity and other areas of IT.
Artificial intelligence is perhaps already a somewhat annoying "buzzword", which, on the other hand, does not mean that we should underestimate it and pretend that it does not concern us. Artificial intelligence can be found in mobile phones, during internet searches, in marketing, in accounting, in transport, and even in such a seemingly unlikely sector as agriculture.
In the following lines, we will focus on the use of artificial intelligence in ICT. But before we get into that, let's first define what AI actually is (and also what it is not).
→ Tip: Take a look at how AI can be used in accounting and controlling.
In very general terms, it is a form of machine intelligence, distinct from the intelligence of animals and humans. It should not - at least for the time being - be confused with the term "general artificial intelligence". It is characterised by the ability to "think" in a completely universal way, and thus the ability to solve any problem (similar to humans). Such a general artificial intelligence does not yet exist, although huge resources are being invested in its development and it is optimistically estimated that its creation is a matter of the next two decades.
Current AI models are capable of simulating human intelligence - but only in a highly specialised area. In some such areas, AI has long since surpassed human intelligence - for example, in playing chess or go. It should be noted, however, that a neural network that can beat even chess grandmasters "with its finger up its nose" will certainly not be able to handle any other problem - for example, a request for text translation or a query about stock market trends. That, on the other hand, can be handled by other AI models (which, on the other hand, could easily beat a small child at chess or go).
We've covered the basics, let's get down to the practical applications of AI in IT.
One of the most interesting uses of AI in software development is code generation. Nowadays, even for free, you can use tools and frameworks that can generate functional code in the desired programming language based on a given specification. An example is probably the most famous chat and generative AI of today - OpenAI GPT-3.5
For example, we tried a request to create a function to sort an array in the Python programming language. The result was displayed in a fraction of a second:
def bubble_sort(arr):
n = len(arr)
for i in range(n):
for j in range(0, n-i-1):
if arr[j] > arr[j+1]:
arr[j], arr[j+1] = arr[j+1], arr[j]
return arr
Another example - we have a task to create a simple Python website with a form for user registration. The chatbot returned the desired output again almost instantly:
from flask import Flask, request, jsonify
app = Flask(__name__)
@app.route('/register', methods=['POST'])
def register_user():
data = request.get_json()
username = data.get('username')
password = data.get('password')
# Validation and saving the user to the database
response = {'message': 'User successfully registered'}
return jsonify(response), 200
if __name__ == '__main__':
app.run(debug=True)
AI can therefore generate code quickly, theoretically saving developers time. Why only in theory? The results need to be verified. Nor can we - yet - expect to create more complex programs. To get a better idea, try having a chatbot write an article on a given topic. The result will be a "plus/minus" match to your assignment, but it will also look clumsy, inelegant, flat, too general - in short, a text full of platitudes and meaningless phrases. Moreover, the facts given will often not correspond to reality. In less sophisticated terms, the AI will simply "suck them off the finger" (or rather the chip).
So think of generative AI as a tool rather than an automaton that does all the work for you. It hasn't taken on much creativity and imagination yet. Of course, in a few months or years the situation may be radically different.
We've already mentioned AI in mobile, but did you know that NLP systems based on AI and ML can also be deployed very successfully in a call centre like this? For example, IBM's system can automatically translate more than 90% of the spoken sentences it "hears" from customers. According to this year's statistics from the infographicjournal website, which talks about the benefits of AI in various telecom industries, call center operators working with Al have seen up to a 61% increase in customer satisfaction.
As cyber threats become increasingly sophisticated, there are growing demands for effective mechanisms to detect and prevent them. AI can analyse large amounts of data and identify unusual patterns indicating a possible attack, for example, detecting suspicious activity on a network and then immediately alerting security teams, or automatically isolating a compromised system or stopping malicious traffic while security teams work to resolve the situation.
Is there a risk of false positives? Sure - we can't completely rule it out. Still, the importance of AI in cybersecurity will grow if only because AI is being used by cybercriminals and hackers themselves.
→ Tip: Check out Sophos cybersecurity products that use AI for heuristic scans in next-gen firewalling as well as for anti-phishing protection.
Artificial intelligence is probably not leaving us. Where its development will take it, no one would dare to say, but what is certain is that it will change our lives. But AI is not just the future, it is the reality of today. At Algotech, we are well aware of this, which is why we already offer products that use AI very effectively - for example, our in-house contact center solution, Algotech CC with advanced voicebots and chatbots.
We are implementing one of the world's most popular CRM systems, SugarCRM - the first ever CRM with datamining AI that was specifically designed to solve sales and marketing challenges. Contact us, we're ready for the future today.
Copyright © 2024 - Algotech a.s., all rights reserved
| Personal data processing |
Terms and conditions
