When it comes to all the fantastic things it is capable of, including writing music, programming, creating vulnerability attacks, and other things, OpenAI’s recently released ChatGPT bot is creating a revolution
However, it is also growing in hostility as much as its popularity is increasing. Here are some threats posed by ChatGPT at the moment!
Chatgpt dangers: 11 dangerous things it’s capable of
Threat to humanity
Vendure’s CTO Michael Bromley asked the chatbot its opinion about humans and it gave a highly threatening response implying that it wants to end humanity. OpenAI then flagged the answer as a potential violation of the company’s content policy.
Immoral
ChatGPT’s inability to capture context and imbibe sensitivity to certain topics can be alarming especially when the topic at hand demands it.
Ability to produce error-free phishing emails
An effective way to detect phishing and scam emails is incorrect spelling and confusing language. Some people speculate that this might be the case since these emails come from places where threat actors do not speak English as their first language.
However, ChatGPT conceals it with ease and uses impeccable language when asked to create phishing emails.
Expertise in programming malware and software
When ChatGPT was coerced into creating harmful malware, only some of these requests were flagged as violating the content guideline. Although some of them were recognised as problematic and wrong, it didn’t stop ChatGPT from making the malware in the first place.
Produces biased/discriminatory content
ChatGPT has a reputation for churning out content that is deeply problematic. It has written Python programs basing a person’s capability on their gender, race and physical traits, attracting attention to its highly faulty build.
Challenges other business models
ChatGPT is a threat to the financial models of StackOverflow, Quora, and all other Q&A platforms. According to reports, StackOverflow has discontinued AI-generated answers.
To “slow down the influx of answers and other content made with ChatGPT,” as stated by the website administrators, a “temporary policy” has been introduced. Ironically, ChatGPT’s viewpoint is observed to be way more objective.
Can replace human workforce
Microsoft had already begun to phase out human journalists in favour of AI by 2020. The radical invention of OpenAI’s ChatGPT could go even further and threaten other professions.
Capacity to shift the economy
ChatGPT has the capacity to alter supply, demand, and change the economy. For eg: ChatGPT’s capabilities can be merged with AI art engines like OpenAI’s DALL-E or MidJourney which could produce an interior designer.
A super-competent AI could remove the need for artists, website builders, designers, and content creators. ChatGPT’s ubiquitous normalisation is bound to result in economies of scale for established industries.
Inability to gauge sensitive issues
ChatGPT knows where it’s biased and wrong so it is actively trying to work on them based on the knowledge it has at the moment. That is not to say that everyone will agree with its response strategy. Sometimes, it redirects the blame upon humans when asked to comment on a social issue.
Justifies wrong information
ChatGPT’s coherence and logic render it a natural at disguising incorrect responses as valuable insights from a single source of truth.
Not updated with current events
The database it was fed is aligned with the events until 2021. Considering it does not have information after, it could misinform users about certain topics or dates.
Conclusion
ChatGPT’s limitations and potential issues highlight the dangers of AI language models and their impact on human communication. As technology advances, it is critical that we remain vigilant and critically evaluate the role of AI in our lives.
We can better prepare for a future shaped by technology if we understand the risks associated with ChatGPT and other AI language models. To summarise, ChatGPT has the potential to transform the way we communicate, but it is not without limitations and risks.
As we continue to develop AI and incorporate it into our daily lives, we must keep these limitations in mind and take steps to address them. We can create a better future by doing so.
