Komal Joshi · Oct 30, 2020 . 7min read
GPT-3 in Blockchain, fuels discussions on Twitter by Vitalik Buterin
Vitalik Buterin's tweet has powered discussion relating to the GPT-3 technology in the Blockchain Space. Individuals argue about the pros and cons of same.
By Shilika · Jul 18, 2020 . 7min read
In June 2020, earlier this year, Open AI announced the release of its paper describing the latest developments in the GPT-3, a language model built of 175 billion parameters with 96 attention layers. Earlier today, Vitalik Buterin, the co-founder of Ethereum, predicted that he could easily see the human workforce replaced by AI builds and human debugs in the coming decades. The immediate application of the same will lead to a compromise in many jobs across different industries and sectors.
The tweet came in response to Sharif Shammen’s statement, founder of Debuild.co. It is an application that provides faster ways to build apps. “I just built a *functioning* React app by describing what I wanted to GPT-3. I’m still in awe”, Sharif tweeted. This was followed by a couple of tweets explaining how his company is leveraging the tool to build apps using plain English.
“I built a todo list app simply by describing it to GPT-3. It generated the React code for a fully functioning app within seconds. I’m becoming more impressed and aware of its capabilities every single day,” he said. Hence, this led to fuelling discussions on Twitter about the same.
What is GPT?
GPT-3 is a giant language model that has a capability of performing NLP (Natural language processing) tasks which it has never seen or known. Therefore, it can generate language in a probabilistic way. If the language model is trained with enough data, then it can solve problems out of the box. Hence it can be widely used to overcome difficulties without human bugs.
Natural language processing (NLP) tasks include translation, summarization, entity recognization, topic segmentation, speech recognization and many more. Datasets are used to filter down the quality and train language models. Common crawl data set along with Web Text 2, Books1, Books2 and Wikipedia is used to train the language model. Common Crawl data set is the crawl of the entire internet to feed in the initial data for the language model.
Role of GPT-3 in Blockchain
The immediate application of GPT-3 in Ethereum can empower people to write smart contracts in standard language. Though challenging, it works then the exceptions could be quite dangerous. The GPT-3 language model has both benefits and flaws concerning its application. In case of misuse, it can lead to dire consequences. Some of them include:
- Writing fraudulent smart contracts
- Abuse of legal and government procedures
- Malicious attacks
The Blockchain space has profoundly been affected due to the ongoing malicious attacks, crimes and illegal activities associated. The latest one reported being the Twitter hack. Moreover, the inappropriate use of GPT-3 can possess a more significant threat to the blockchain space, thus raising ongoing security concerns. The Blockchain tech is still developing. Hence the lack of security infrastructure can lead to some individuals menacing the area.
Follow Cryptodose for more updates
Komal Joshi · Oct 30, 2020 . 5min read