Example: If a facial recognition system is trained mostly on lighter-skinned faces, it might not perform well for people with darker skin tones.
Since deep learning needs large datasets, it often uses personal information, which can raise privacy concerns.
Why it’s a problem:
Collecting personal data without permission is unethical and sometimes illegal.
Biased or unfair data can lead to decisions that harm certain groups of people.
Example: AI used in hiring has been found to favor men over women if the training data austria telegram data included mostly male applicants.
8. Hard to Use in Real Life
Deploying deep learning models into real-world systems can be tricky. They often need to be adjusted to work efficiently on different devices, like mobile phones or cloud systems.
Why it’s a problem:
Ensuring the model performs well in all situations can take time and effort.
If the environment changes (e.g., weather for self-driving cars), the model might struggle.
- Board index
- All times are UTC
- Delete cookies
- Contact us