On March 21, 2023, Google released its highly anticipated AI Chatbot called Bard.
Initially, it was based on the Large Language Model, and the upgraded version is based on PaLM or Pathways Language Model.
Long before it was released, in October 2022, Google Bard was blocked due to some concerns relating to the technology.
This article gives you an in-depth view of why Bard was blocked and how it returned to reality.
Table of Contents Show
Introduction To Google Bard
Similar to GPT, Bard is a conversation AI, and according to Google, it is more user-friendly.
It gives human-like answers and is more logical and up-to-date than its competitors.
Furthermore, in October, Google Bard was blocked for all the wrong reasons raising concerns among people.
Experts speculated that the model could generate hate speech or misinformation. Another concern was its ability to imitate a real person.
The internet will be an unethical nuisance for them, and AI will further fuel the already blazing volcano of hate and chaos.
Reasons Why Google Bard Was Blocked
Some of the major reasons why Google Bard was blocked were its ability to create harmful content and incomprehension.
- Harmful Content: The Dataset upon which the AI was trained contained harmful material, and the model could generate harmful content by itself.
- Incomprehension: Like Humans, the AI could not comprehend between good and evil eventually, ignoring the harmful effect of immoral, unethical content.
The AI-generated content was harmful even when not told to do so. Google Bard itself states why it was blocked before its release.
Measures Taken By Google To Get Bard Unblocked
Since the backlash in October, Google has implemented numerous measures and steps to mitigate such concerns in the future.
After all, using AI for unethical purposes can pose a significant threat to people.
Here are some measures Google took to unblock Bard;
1. Detection Of Harmful Content
Google claims to have improved the AI’s ability to detect and avoid harmful content or prompts a user enters.
After inputting inappropriate prompts, the AI simply replies, “I am just a language model, so I can’t help you with it.”
2. Distinguishing Harmful And Non-Harmful Content
Google has trained its AI rigorously to make it understand the difference between right and wrong.
It was trained on a massive data set, developing algorithms to detect harmful content and working with experts.
3. Transparency
To increase transparency with the users, Google has released papers about how their model functions.
This allows the users to understand how Bard AI works and the terms of service users must follow.
The Bottom Line
Concerns over Artificial Intelligence are also rising with the increased level of development in the industry.
The effect of harmful content is a serious threat. Companies like OpenAI and Google are responsible for promoting clean AI culture without harming humans.
Google Bard in October 2022 was also blocked upon a similar pretext.
Furthermore, Google was forced to delay the release of the AI until the Company improved Bard’s ability to detect, distinguish, and avoid harmful content and prompts.
Finally, in March 2023, the now-popular Bard AI was released.
