by ChatGPT is a chatbot powered by artificial intelligence (AI), developed by OpenAI, which has gained great popularity among netizens as a service that allows you to make inquiries on any topic and with an interaction similar to a conversation with a human being.
For this reason, this chatbot is being used as a means to generate questions about different issues of history, politics, science, and even certain users are turning to this tool to obtain advice on finances, cooking recipes and even tips to get a partner in dating services on-line like tinder.
Thanks to the fact that the system was trained by analyzing large amounts of text that is published on millions of web pages, the chatbot has the ability to use the information it has learned to answer almost any question in a fluid way.
However, cybersecurity experts have exposed that this technology also presents certain risks in terms of computer security. According to Sol González, Computer Security Researcher at ESET Latin America, this system can be used by cybercriminals as a tool to generate malicious code and carry out other types of fraudulent actions.
How do cybercriminals take advantage of ChatGPT to attack people?
phishing attacks
Criminals can use the chatbot as a tool to generate malicious emails that are much more persuasive, since artificial intelligence has the ability to generate a text with fewer misspellings and better consistency than the malicious emails people often receive.
This situation would make fake emails that impersonate recognized brands more credible by being better worded.
So users should be much more cautious when reviewing emails that seem suspicious, therefore, it is key not to enter the links that accompany these communications, or download the attached files they have. Besides, you should always consult the official communication channels of the brands, especially in the case of banks, in case any problem is reported with the account.
Identity Theft
As with malicious emails, cybercriminals can use artificial intelligence to create scams that attempt to impersonate well-known brands (banks, online platforms, etc.). streamingwarehouses, services on-line). In this way, criminals try to steal personal information from victims to access their accounts.
Create malicious code
This platform can be used for software development, since it can be used as a tool to generate code in various programming languages. Despite the fact that ChatGPT warns about the risks of this type of request and points out that ethical standards could be violated, along with the service’s policy, some users have managed to obtain that code along with a detailed explanation of how it works.
malicious chat
ChatGPT has an API that allows it to feed other chats, thanks to which this technology can be used in other messaging platforms, such as WhatsApp, but it can also be used for malicious purposes. For example, generating chats to scam people using more credible information.
Automate processes in cyberattacks
Before proceeding with an attack, cybercriminals typically run a reconnaissance process that incorporates several repetitive tasks. However, ChatGPT could take care of those activities, making it easier to design and implement a hack against an entity.
What is ChatGPT?
This service represents an evolution of the artificial intelligence model called GPT-3, developed by OpenAI, and which has positioned itself as one of the most advanced AIs today.
Artificial intelligence belongs to the group of systems called Large Language Models (large language models) because they are based on machine learning technology, which gives them the ability to read, summarize and translate texts.
In addition, they have the ability to predict future words in a sentence, a function that gives them the fluency in producing texts with a structure similar to that used by humans when they speak or write.
Added to this, this technology has been trained by analyzing large amounts of texts that are published on millions of web pages, including forums, news portals, academic websites, among other sites.
How does ChatGPT work?
It is enough to enter the official website of the platform and register with an email and phone number, then the user can open a conversation to interact with the chatbot.
Within a conversation, the user can make all kinds of queries, including being asked to explain the concept of a complex text in a simpler way, that generates song lyrics or tells a joke and it is even possible to establish a guessing game about TV seriessongs or video games based on emoticons.
