{"id":3679,"date":"2025-04-20T17:37:01","date_gmt":"2025-04-20T17:37:01","guid":{"rendered":"https:\/\/jornaloassobio.pt\/?p=3679"},"modified":"2025-04-20T17:37:01","modified_gmt":"2025-04-20T17:37:01","slug":"inteligencia-artificial-texto-de-opiniao","status":"publish","type":"post","link":"https:\/\/jornaloassobio.pt\/?p=3679","title":{"rendered":"Intelig\u00eancia Artificial – texto de opini\u00e3o"},"content":{"rendered":"
\r\n
\r\n \t<\/i> Read Time:<\/span>2 Minute, 11 Second <\/div>\r\n\r\n <\/div>

\"\"<\/p>\n

Texto de opini\u00e3o sobre Intelig\u00eancia Artificial, escrito pelo aluno Duarte Jos\u00e9 Pat\u00e3o S\u00e3o Pedro.<\/p>\n

 <\/p>\n

In the past years, AI technology has grown, expanded and evolved at an impressive, and maybe scary rate. As of today, anyone can visit OpenAI\u2019s ChatGPT, or its newer, slightly controversial Chinese cousin DeepSeek, and ask for any information, tutorial, or even generated images.<\/p>\n

Of course AI models do have their limitations and bounds set by their respective creators, and while being called artificial intelligence, they don\u2019t do their name justice. ChatGPT is infamous for acknowledging its mistakes, but repeating them over and over.<\/p>\n

Still, when dealing with simple and repetitive tasks, AI can be a very sharp and effective tool. Monotonous, boring and repetitive work is where it shines the most.<\/p>\n

In surveillance, there is potential for AI models on facial recognition and automatically managed databases. In the military, there\u2019s target recognition, navigation, and faster, plus more accurate automated ballistic calculation. In the area of transport, AI could be used on autopilot, self-driving and security systems\u2026 AI is everywhere already. Even in your own home, if you have Siri or Alexa systems. EVEN in your nearest MacDonald\u2019s, where AI drive-thru ordering is being implemented.<\/p>\n

We are lazy creatures who like having things done for us by other lazy creatures. And AI \u2013 plus technology in general \u2013 does exactly that. All of these advantages make tech companies develop better and better products \u2013 in this case, ever-progressing AI models.<\/p>\n

However, of course there is a catch. It\u2019s not all as great as it sounds. First of all, AI makes mistakes. What could be a small meaningless grammar error, for the average sleepless college student who just wants to generate an essay of sorts, could be a deadly mistake for a pilot who decided to fully trust the autopilot system and take a nap. Also, AI can sometimes flee its rules and provide unethical information to the wrong people, which could end badly. For the worst part, governments of the world might use (and might already be using) AI for abusive and invasive surveillance, risking the privacy of the citizen, China being one of the worst examples of this\u2026 (Not so fun fact: when asked about anything related to the Tiananmen Square protests, DeepSeek censors itself mid-generation \u2013 Can\u2019t imagine why that is\u2026)<\/p>\n

\u201cHackers might not be the threat, but salvation itself.\u201d<\/p>\n

A (slightly adapted) quote from WATCH_DOGS 2, A movie-worthy game that portrays a dystopian reality that is getting closer and closer to ours.<\/p>\n

 <\/p>\n