Page 44 - Touhpad Ai

u To improve accuracy and efficiency: AI can make faster and more accurate decisions than humans when large
amounts of data are involved.
Example: Fraud detection in banking, where AI scans millions of transactions in seconds to flag suspicious activity.
u To personalise experiences: AI decision making allows systems to adapt to individual users.
Example: Netflix or YouTube deciding what to recommend based on your viewing history.
u To reduce human effort: Machines with AI decision making can take over repetitive or risky tasks, saving time and
protecting humans.
Example: Robots in factories deciding the order of assembly or welding without human supervision.
u To enable learning and improvement: Unlike simple machines, AI can learn from past decisions and improve
over time.
Example: Voice assistants like Alexa become better at recognising your voice commands after repeated use.
u To support human decision making: AI doesn’t always replace humans; sometimes, it helps them make better
choices.
Example: Doctors use AI-based systems for medical image analysis to confirm diagnoses.

Cyber Security in Computing and Machine Intelligence
Cyber security is the process of protecting computers, networks, software, and important data from being damaged,
stolen, or accessed by unauthorised individuals. You can think of cyber security as a digital security guard that keeps your
personal information safe, prevents others from misusing systems, and ensures that everything works as it should. Just
like we lock our homes to stay safe, computers and AI systems also need protection from online threats.
As Artificial Intelligence (AI) systems become smarter and more common in daily life, they
are now used to handle important tasks such as managing bank transactions, storing
patient medical records, controlling traffic signals, recognizing faces and voices, and
even teaching students through online platforms. Since these AI systems handle sensitive
and valuable information, it is extremely important to protect them. If AI systems are
not protected, hackers can break into them and steal or change information. Worse, if
attackers feed incorrect or malicious data into an AI system—a process called data
poisoning—it may start making inaccurate or harmful decisions. This can lead to serious
problems, such as a health app giving wrong advice or a security camera failing to detect
danger. AI tools can also be misused to create fake videos, voices, or news content, known as deepfakes, which can fool
people and spread false information. That’s why robust cyber security measures are necessary to keep AI systems safe
and trustworthy.
There are many types of cyber threats that AI systems face. Few are listed below:

Threat Description
Data Theft Someone steals private data like passwords, health info, or bank details.

Phishing Attacks Fake emails or messages try to trick users into giving secret information.
Data Poisoning Attackers feed wrong data into the AI system so that it learns incorrectly.

Deepfakes AI-generated fake videos or audio used to mislead people.
Hacking An attacker takes control of the system and changes how it works.

To protect AI systems from these threats, we use various cyber security techniques:
u Encryption which hides your data by turning it into secret code that only authorized users can understand.
u Two-factor authentication (2FA) adds an extra step to verify your identity, such as entering a password plus a
one-time password (OTP).

42 Touchpad Artificial Intelligence - XI