The roaring success of ChatGPT has triggered a race among Big Tech companies to either come up with their own competing artificial intelligence products or buy a stake in promising startups that could become the next OpenAI, the creator of ChatGPT. In the past month, Microsoft, Google and China's Baidu have each presented their responses to ChatGPT. In the meantime, some of these companies have also invested or formed partnerships with lesser-known startups specializing in generative A.I., the technology behind text and image generators like ChatGPT and Dall-E, OpenAI's other viral product that can generate digital images based on text prompts. There is a natural attraction between A.I. startups and tech behemoths. Many startups rely on the cloud infrastructure of large tech companies to train their algorithms, while tech giants often see them as potential investment or acquisition targets to expand their business without having to do the early-stage research themselves.

ChatGPT has revolutionized the way we interact with AI. Its advanced natural language processing capabilities have made it possible for businesses and individuals to automate their communication processes in a way that is both efficient and effective. This has not only made it easier for companies to provide high-quality customer service, but it has also made it possible for individuals to create content quickly and efficiently. More podcasts: Podcasting, Online Learning & Innovation ChatGPT is revolutionizing the way we engage with artificial intelligence. An AI-powered ChatBot can have natural-sounding conversations with users. It is used to comprehend queries and give precise answers.

When Microsoft released Bing Chat, an AI-powered chatbot co-developed with OpenAI, it didn't take long before users found creative ways to break it. Using carefully tailored inputs, users were able to get it to profess love, threaten harm, defend the Holocaust and invent conspiracy theories. Can AI ever be protected from these malicious prompts? What set it off is malicious prompt engineering, or when an AI, like Bing Chat, that uses text-based instructions -- prompts -- to accomplish tasks is tricked by malicious, adversarial prompts (e.g. to perform tasks that weren't a part of its objective. Bing Chat wasn't designed with the intention of writing neo-Nazi propaganda.

Microsoft and OpenAI have partnered up to explore the possibility of using the ChatGPT language model to control robots and drones using natural language. By doing so, the companies aim to simplify interactions between people and machines, without the need for complex programming languages. The goal is to enable ChatGPT to control robots, making it easier for people to interact with them, and improve communication through the implementation of artificial intelligence developed by OpenAI. Microsoft has released a paper that outlines a new set of design principles that utilizes ChatGPT to give instructions to robots. The process involves defining a list of high-level tasks that the robot can perform, writing an instruction that ChatGPT translates into the robot's language, and then running a simulation in which the robot follows the instructions.

Is conversational AI a blessing or curse for data? If you follow the tech industry, you’ve heard about ChatGPT. Whether you think it’s the future of chatbot technology or you’re erring on the side of caution, if you know about it, you’re bound to have an opinion. As Google confirms it’s launching a rivalling service, interacting with AI will soon become commonplace in our personal and professional lives. But what does that mean for data and analytics? Here, Jonathan Hedger, co-founder of the UK’s only data jobs board, Only Data Jobs, explores. Launched late in 2022, ChatGPT has quickly become

The Coca-Cola Company is getting deep into the artificial intelligence game thanks to a newly-inked deal that will see it partnering with OpenAI. The press release announcing the deal did not say how much the partnership between Coca-Cola, OpenAI, and the Bain consulting firm is worth, but given that the AI firm recently got a multi-billion-dollar contract with Microsoft, there's a good chance it's worth a pretty penny -- and it seems particularly significant because of Coke's storied history in advertising, a domain that the release seems to hint could be a target for the AI tech. In the statement, Coca-Cola CEO James Quincy said that the company is "excited to unleash the next generation of creativity offered by this rapidly emerging technology" using tech including DALL-E and ChatGPT. "We see opportunities to enhance our marketing through cutting-edge AI," Quincy said, "along with exploring ways to improve our business operations and capabilities." Coke is the first company to sign on to Bain and OpenAI's new partnership, and although there's no word yet on exactly what the deal will entail, the press release hints at marketing, sales, and human resources implementations, the latter two of which sound pretty dystopian, to be honest.

Pre-trained models (PM) have achieved promising results in content generation. However, the space for human creativity and imagination is endless, and it is still unclear whether the existing models can meet the needs. Model-generated content faces uncontrollable responsibility and potential unethical problems. This paper presents the MetaAID 2.0 framework, dedicated to human-controllable PM information flow. Through the PM information flow, humans can autonomously control their creativity. Through the Universal Resource Identifier extension (URI-extension), the responsibility of the model outputs can be controlled. Our framework includes modules for handling multimodal data and supporting transformation and generation. The URI-extension consists of URI, detailed description, and URI embeddings, and supports fuzzy retrieval of model outputs. Based on this framework, we conduct experiments on PM information flow and URI embeddings, and the results demonstrate the good performance of our system.

Natural language processing based on large language models (LLMs) is a booming field of AI research. After neural networks have proven to outperform humans in games and practical domains based on pattern recognition, we might stand now at a road junction where artificial entities might eventually enter the realm of human communication. However, this comes with serious risks. Due to the inherent limitations regarding the reliability of neural networks, overreliance on LLMs can have disruptive consequences. Since it will be increasingly difficult to distinguish between human-written and machine-generated text, one is confronted with new ethical challenges. This begins with the no longer undoubtedly verifiable human authorship and continues with various types of fraud, such as a new form of plagiarism. This also concerns the violation of privacy rights, the possibility of circulating counterfeits of humans, and, last but not least, it makes a massive spread of misinformation possible.

Schema induction builds a graph representation explaining how events unfold in a scenario. Existing approaches have been based on information retrieval (IR) and information extraction(IE), often with limited human curation. We demonstrate a human-in-the-loop schema induction system powered by GPT-3. We first describe the different modules of our system, including prompting to generate schematic elements, manual edit of those elements, and conversion of those into a schema graph. By qualitatively comparing our system to previous ones, we show that our system not only transfers to new domains more easily than previous approaches, but also reduces efforts of human curation thanks to our interactive interface.

Human language is grounded on multimodal knowledge including visual knowledge like colors, sizes, and shapes. However, current large-scale pre-trained language models rely on text-only self-supervised training with massive text data, which precludes them from utilizing relevant visual information when necessary. LM, to Visually-augment text tokens with retrieved relevant images for Language Modeling. LM builds on a novel latent text-image alignment method via an image retrieval module to fetch corresponding images given a textual context. LM uses a visual knowledge fusion layer to enable multimodal grounded language modeling by attending to both text context and visual knowledge in images. LM on various visual knowledge-intensive commonsense reasoning tasks, which require visual information to excel. LM outperforms all strong language-only and vision-language baselines with substantial gains in reasoning object commonsense including color, size, and shape. Our code is available at https://github.com/Victorwz/VaLM. Large-scale pre-trained language models (PLMs) have achieved great success in promoting state of the art on various natural language understanding and generation tasks (Devlin et al., 2019; Radford et al., 2019; Liu et al., 2019; Yang et al., 2019; Brown et al., 2020; Wang et al., 2022). PLM self-supervision training largely benefits from harvesting local context information in the pre-training corpus.