Social media is awash with people talking about a new piece of software called GPT-3, which has been developed by OpenAI, an Elon Musk-backed artificial intelligence lab in San Francisco. GPT-3 (Generative Pre-training) is a language-generation tool capable of producing human-like text on demand. The software learned how to produce text by analyzing vast quantities on the internet and observing which letters and words tend to follow one another. OpenAI started releasing it to a select few people last week who had requested access to a private early version, and many of them have been blown away. "It's far more coherent than any AI language system I've ever tried," wrote entrepreneur Arram Sabeti in a blog post after testing. "All you have to do is write a prompt and it'll add text it thinks would plausibly follow.

In July, San Francisco-based OpenAI stunned everyone with the capabilities of its AI language model, the GPT-3. The text generator can pen fiction, compose poetry and write business memos — all without any human intervention. Simply put, it is being seen as a tool that brings machines a wee bit closer to mimicking human intelligence. A look at how AI has come this far.The GPT-3, or Generative Pre-training Transformer 3, can use half a sentence as input and type out the rest correctly. It doesn't stop at that. The text predictor can then type out a whole para, or a book, that make logical sense. The GPT-3, however, lacks the ability to reason abstractly. It is at a loss when faced with new ideas. ​First, what is the big fuss really?

Last spring, artificial intelligence research institute OpenAI said it had made software so good at generating text--including fake news articles--that it was too dangerous to release. That line in the sand was soon erased when two recent master's grads recreated the software and OpenAI released the original, saying awareness of the risks had grown and it hadn't seen evidence of misuse. Now the lab is back with a more powerful text generator and a new pitch: Pay us to put it to work in your business. Thursday, OpenAI launched a cloud service that a handful of companies are already using to improve search or provide feedback on answers to math problems. OpenAI was founded as a nonprofit in 2015 by Elon Musk and other Silicon Valley notables to ensure that future superhuman AI was a benign force.

I've been playing around with OpenAI's new GPT-3 language model. When I got beta access, the first thing I wondered was, how human is GPT-3? How close is it to passing a Turing test? Let me explain how exactly I'm generating these conversations. GPT-3 is a general language model, trained on a large amount of uncategorized text from the internet.

I was wondering if there are any on-going efforts to build a database such as openwebtext but for academic pre-print repositories such as arxiv and ssrn? Of course, having final versions of papers as published in journals or conferences would be best but that may prove harder to get. And then again, most authors always put the final version of the paper on the pre-print websites. I was thinking about how having a model such as GPT-3 but trained on domain knowledge from the above-mentioned pre-print sites, can help surface deep connections during the writing process. Imagine giving GPT-4 a latex code for your table and having it produce a discussion of the results and drawing insights between your numbers and other similar numbers as reported in the literature.

We will use Google Drive to save our checkpoints (a checkpoint is our last saved trained model). Once our trained model is saved we can load it whenever we want to generate both conditional and unconditional texts. Now that you have your Google Drive connected let's create a checkpoints folder: Now let's clone the GPT-2 repository that we will use, which is forked from nnsheperd's awesome repository (which is forked from OpenAI's but with the awesome addition of train.py), I have added a conditional_model() method which will let us pass multiple sentences at once and return a dictionary with the relevant model output samples. It also lets us avoid using bash-code.

Let's start with the basics. GPT-3 stands for Generative Pretrained Transformer version 3, and it is a sequence transduction model. Simply put, sequence transduction is a technique that transforms an input sequence to an output sequence. GPT-3 is a language model, which means that, using sequence transduction, it can predict the likelihood of an output sequence given an input sequence. This can be used, for instance to predict which word makes the most sense given a text sequence.

A lot of ink has been spilled (or pixels illuminated) about the wonders of GPT-3, OpenAI's latest and greatest language model. A team of more than 30 OpenAI researchers have released a paper about GPT-3, a language model capable of achieving state-of-the-art results on a set of benchmark and unique natural language processing tasks that range from language translation to generating news articles to answering SAT questions. But like most examples spat out by language models, almost all of these were hand-selected by humans after many runs. Because not-so-good results just wouldn't make the news. Even bearing that in mind, I'm still blown away by what I've seen of GPT-3.

Did you know AI can now produce poetry and write fiction? That is not all it can also generate CODE ! Isn’t that amazing ! OpenAI’s newest AI language model – GPT-3, is trending all over the internet! Source: Unite.AI Background Elon Musk and Sam Altman started OpenAI in 2015 to advance the state of the art of AI and to ensure AI was used for the human good. OpenAI recently released the third version Generative Pre-training Transformer (GPT) – GPT-3. They first described GPT-3 through a research paper published in May but last week rolled out a beta version access to a select set of people and its capabilities are mind blowing! What is GPT-3 and why is everyone talking about it? Here are some quick pointers to help you understand what GPT-3 is, and why it is a step towards an AI revolution. Third version Generative Pre-training Transformer (GPT) A natural language generator (NLG) capable of producing human-like text on demand State-of-the-art language model made up of 175 billion parameters ~10X larger than Microsoft’s Turing NLG that has 17 billion parameters >100X larger than its own predecessor GPT-2 (released last year) trained on 1.5 billion parameters Does not require large custom, task specific datasets (which are usually difficult to get) Does not even require task specific model architectures What all can GPT-3 do for you? Here are a few things that GPT-3 can do for you and as more developers and experts experiment with it, more use cases will surface in the times to come. Answer questions with common sense (that doesn’t seem very difficult?) Write creative fiction – poetry, essays, stories Write news (Could that be a problem?) Solve arithmetic problems Generate Functioning Code (GPT-3 can CODE!) Design – developers demonstrated it with a Figma plugin (That’s truly creative!) This is just an illustrative list of what all GPT-3 can do! As more and more developers/experts experiment with the beta access, more innovations are bound to surface. Possible flaws and concerns pertaining to GPT-3? Although GPT-3 has proven brilliance in many ways in its current state but it also has certain flaws and concerns that it raises: Lacks an overarching, long-term sense of meaning and purpose As it generates its output word-by-word, based on the immediately surrounding text. It can struggle to maintain a coherent narrative or deliver a meaningful message over more than a few paragraphs is what experts say after initial experimentation. Possibility of being prone to certain biases Developers noticed that GPT-3 is prone to shoot out racist and sexist language, even when the prompt is something harmless. As GPT-3 is trained on internet scale data, these biases arise from biases in that training data reflecting possible societal views and opinions. Chances of misuse once released publicly Authors acknowledge that people can misuse it in several ways because of its ability to create text that is indistinguishable from that of written by humans. It can lead to creation of including generating misinformation and spam, phishing, and even fake academic essays. Possible impact on jobs As GPT-3 can perform human-like tasks across multiple domains, it could possibly have an impact on jobs that it can do like writers, coders, journalists, etc. These are just possible concerns and basis point 1 highlighted above many believe that instead of replacing humans, GPT-3 could become the perfect assistant for humans in these professions. Conclusion and what lies ahead? GPT-3 can execute plethora of NLP- based tasks, without fine-tuning for a specific task. Experts are even saying that this could be a step towards Artificial General Intelligence (AGI) – Read my article on Demystifying AI to know more! It is capable of performing machine translation, answering to questions, scripting poems and stories, elementary mathematics and can even generate code. OpenAI wants developers to help it explore what GPT-3 can do and with the beta release of its API has attracted plethora of experts and developers to experiment with the capabilities of GPT-3. It is only a matter of time before more exciting innovations surface in the developers’ community! Watch out for more interesting articles on AI! Feel free to share your thoughts. References [1] https://www.forbes.com/sites/robtoews/2020/07/19/gpt-3-is-amazingand-overhyped/#7c6166501b1c [2] https://analyticsindiamag.com/open-ai-gpt-3-code-generator-app-building/ [3] https://singularityhub.com/2020/06/18/openais-new-text-generator-writes-even-more-like-a-human/ [4] https://towardsdatascience.com/gpt-3-for-the-people-2cdd003d9a89 [5] https://www.technologyreview.com/2020/07/20/1005454/openai-machine-learning-language-generator-gpt-3-nlp/ [6] https://www.analyticssteps.com/blogs/what-openai-gpt-3 [7] https://www.cnbc.com/2020/07/23/openai-gpt3-explainer.html [8] https://analyticsindiamag.com/how-openais-gpt-3-can-be-alarming-for-the-society/ [9] https://www.forbes.com/sites/robtoews/2020/07/19/gpt-3-is-amazingand-overhyped/#7c6166501b1c [10] https://www.independent.co.uk/life-style/gadgets-and-tech/news/gpt3-ai-tool-designs-websites-medicine-a9627966.html [11] https://medium.com/fair-bytes/how-biased-is-gpt-3-5b2b91f1177 [12] https://towardsdatascience.com/gpt-3-demos-use-cases-implications-77f86e540dc1

The world has a new AI toy, and it's called GPT-3. The latest iteration of OpenAI's text generating model has left many starstruck by its abilities – although its hype may be too much. GPT-3 is a machine learning system that has been fed 45TB of text data, an unprecedented amount. All that training allows it to generate sorts of written content: stories, code, legal jargon, all based on just a few input words or sentences. And the beta test has already produced some jaw-dropping results.