ChatGPT Rise of the Chatbots:      Good news, disinformation or misinformation at scale?

ChatGPT Rise of the Chatbots: Good news, disinformation or misinformation at scale?

Siri, Alexa, Google search, personal tutoring & much more might never be the same after this. Released recently, ChatGPT took social media by storm & there is a wave of enthusiasm as many got impressed by its potential upon try. In fact during a recent meeting, Google employees also asked executives whether this is a “missed opportunity” or "threat" for their company.

Designed to operate as a digital assistant, ChatGPT retrieves & repackages info with unmatched speed & with very interesting results. Take a look at few of mine where it becomes obvious how much more human consumable its results are.

No alt text provided for this image
Google Results vs ChatGPT as a cook


No alt text provided for this image
ChatGPT as a poet


No alt text provided for this image
ChatGPT as a programmer


No alt text provided for this image
ChatGPT view on AI

This seems to be a game changer as it also has the potential to reinvent & possibly replace traditional search engines soon. As you saw, it not only offers answers in human friendly sentences (i.e. instead of long link lists) but it also manages to communicate concepts in ways that makes sense for humans.

There is however a catch as it’s not always truth that prevails but in some cases fact is blended with fiction & in others the tool can be misused to create spam & confusion. This leads to the dark side of corrupted AI & deceptive agents where the intention is to spread misinformation or even target specific groups with conversations that aim to convince of particular points of view.

As an example, it comes as no surprise that Stack Overflow, a community built upon trust banned the use of ChatGPT for posts as it makes it too easy for users to generate responses & flood the site with answers that seem correct at first but prove to be wrong on close examination. Similarly, such a problem seems to also be unaddressed in Twitter, Reddit, Discord and LinkedIn & demonstrates that a deep fake internet is only starting to emerge.

OpenAI did in fact warn that ChatGPT is currently a research project that “may occasionally generate incorrect information” and/or “produce harmful instructions or biased content.” Although they intend to refine this technology & these are exciting times, well thought, careful steps are certainly needed especially considering the speed that innovation takes place nowadays.

Brian Haddigan

Chief Executive Officer & Co Founder @ Smart Reactors & Cerefuze | Medical Device Coatings

1 年

Keep up the great work Ioannis. Happy Christmas.

回复

要查看或添加评论,请登录

社区洞察

其他会员也浏览了