Research in the field of machine learning and AI, now a key technology in practically every industry and company, is far too voluminous for anyone to read it all. This column aims to collect some of the most relevant recent discoveries and papers -- particularly in, but not limited to, artificial intelligence -- and explain why they matter. This week in AI, scientists conducted a fascinating experiment to predict how "market-driven" platforms like food delivery and ride-hailing businesses affect the overall economy when they're optimized for different objectives, like maximizing revenue. Elsewhere, demonstrating the versatility of AI, a team hailing from ETH Zurich developed a system that can read tree heights from satellite images, while a separate group of researchers tested a system to predict a startup's success from public web data. The market-driven platform work builds on Salesforce's AI Economist, an open source research environment for understanding how AI could improve economic policy.

The industry is not blind to the problem, and this weakness is forcing cybersecurity companies to take a much broader approach to bolstering defenses. One way to help prevent data poisoning is for scientists who develop AI models to regularly check that all the labels in their training data are accurate. OpenAI LLP, the research company co-founded by Elon Musk, said that when its researchers curated their data sets for a new image-generating tool, they would regularly pass the data through special filters to ensure the accuracy of each label. "[That] removes the large majority of images which are falsely labeled," a spokeswoman said.

Artificial intelligence has endowed us with limitless possibilities. From intelligent marketing to fraud prevention and 24/7 customer support, artificial intelligence has transformed every aspect of businesses and lives. Today, it can also enable machines to use textual or visual data to create new content via what we can refer to as Generative AI. Generative AI refers to artificial intelligence algorithms that enable using existing content like text, audio files, or images to create new plausible content. In other words, it allows computers to abstract the underlying pattern related to the input, and then use that to generate similar content.

GPT-3's highest and the most accurate model Davinci costs 6 cents for every 1000 tokens. So it isn't really inexpensive to operate at scale in a production app. So beyond designing prompts, it is essential to even master the craft of smart prompting, that is to reduce the number of tokens in the input prompt. In this tutorial, we will see a few techniques to reduce the number of tokens in a given prompt from my experience of building supermeme.ai, And remember every 1000 tokens reduced is 6-cents (0.06$) saved, so at scale this is huge.

Google demonstrates impressive artificial intelligence image analysis: the multimodal trained LiT model outperforms OpenAI's CLIP. The combination of images and text descriptions, usually pulled en masse from the Internet, has proven to be a powerful resource for artificial intelligence training. Instead of relying on manually crafted image databases like ImageNet, where people search numerous images for each category like dog, cat, or table, newer image analysis models rely on comparatively unstructured masses of images and text. They learn multimodally and self-monitored. A particularly prominent example is OpenAI's CLIP, which is used, for example, in the new DALL-E 2. These self-supervised trained AI models have one major advantage: they learn much more robust representations of visual categories, since they do not have to rely on the categorizations manually identified by humans.

A decade ago, the conventional wisdom was that AI would first impact physical labor, and then cognitive labor, and then maybe someday it could do creative work. It now looks like it's going to go in the opposite order.

When OpenAI's DALL.E 2 was released two weeks back, the AI tool's ability to create images using sparse natural language instructions caused an online frenzy. Whatever its predecessor DALL.E could do, DALL.E 2 could do better. After the announcement, OpenAI's CEO Sam Altman spoke about the potential upsides of DALL.E 2 and the general direction that AI was moving towards in his blog. According to Altman, the general idea that AI's contributions would affect physical labour first, followed by cognitive labour and then eventually reach creative work has been reversed in reality. "It now looks like it's going to go in the opposite order," he noted.

Like many lonely children, Lucas Rizzotto had an imaginary friend: a talking microwave called Magnetron. As the years passed, the pals drifted apart. But Rizzotto never forgot about Magnetron. When OpenAI released the GPT-3 language model, Rizzotto saw a chance to rekindle the friendship. His story provides a cautionary tale about the dangers -- and delights -- of AI.

With AI (artificial intelligence) making significant advancements in recent years, major corporations around the globe are getting more inclined toward investing in speech recognition. The ultimate goal of this particular technology is to be able to communicate, interpret, and generate human-level speech. In 2020, OpenAI unveiled GPT-3, which stunned the world, thanks to its unrivaled human-level language interpretation. Some industry pundits couldn't resist calling the technology'intelligent' and'sentient'. That's not all, as Google unveiled two of its powerful language models, named LaMDA and MUM, in 2021.

Since around 2017, Google has been pushing ahead with its own transformation into the world's leading AI company. The company also wants to implement this internally. "It's important to be honest", OpenAI founder Sam Altman wrote these days in the context of the question of whether artificial intelligence eliminates more jobs than it creates. The occasion was OpenAI's new art AI DALL-E 2, which generates graphics from sentences. Altman estimates that at first the creative work, then the cognitive and finally the physical work can be replaced by artificial intelligence.