- Work With AI
- Posts
- Google's Bard Fumbles in It's First Week
Google's Bard Fumbles in It's First Week
02-09-2023
⚡ Today’s Highlights
📰 News: Alphabet shares plummet after Bard makes a major mistake in demonstration
💰 Funding: Data observability, investment intelligence, and medical AI that spots life-threatening illness in CT scans
🦾 Tools: A generative AI-based tool that can create physical products from written descriptions?!
📅 Events: The World AI Cannes Festival starts today! (February 9-11, 2023. Cannes, France + Virtual)
💼 Jobs: Experian and Bosch
📰 Today's Top Stories
(5 min read) (Source: Reuters)
TLDR: Seems like Alphabet has been scrambling lately… After announcing a CODE RED a few weeks ago and calling founders Larry Page and Sergey Brin for help, shares of Google parent Alphabet dropped heavily yesterday, wiping billions of dollars off its market value, as Wall Street reacted to the back and forth they’ve had with Microsoft this week.
During a live-streamed presentation from Paris, Google demonstrated some of the capabilities of Bard and introduced some cool new features for its product line including Google Lens, Google Maps, and Google Translate. Unfortunately, experts quickly pointed out that Bard made astronomical factual errors, highlighting the larger issue with AI-powered search and the potential for spreading misinformation.
Google seems to be struggling to keep up with Microsoft which is quickly embedding OpenAI’s technologies into its own consumer products. The fumble caused Alphabet Inc to lose $100 billion in market value on Wednesday. Ouch.
The Big Picture: Bing is coming for Google’s spot as the top dog in the search engine game very quickly and Google just keeps fumbling the ball. It seems like Microsoft won this battle, who will win the war?
(13 min read) (Source: The Atlantic)
TLDR: There are millions of people creating and using all sorts of new tools powered by artificial intelligence and natural language processing. Currently, it’s being used primarily to guide image-generating models like DALL-E 2 and text-generation tools like ChatGPT. But AI-based tools operating through natural language will soon be everywhere in the market, from designing clothes to processing gigabytes of data.
Prompt engineering, the ability to instruct/speak to these AI models to produce high-quality results, is a new and rapidly growing field that combines elements of linguistics and problem-solving. A crucial aspect of prompt engineering is iteration and testing, as well as having subject-area expertise and understanding the unique strengths and limitations of each tool.
The most successful prompt engineers possess the ability to create specific, consistent, and predictable results by experimenting with different words and parameters. The field is becoming increasingly important as AI-generated creations become more widespread, and it lies at the intersection of conversation, query, programming, and prose. This combination of technical knowledge and creativity makes prompt engineering a distinctly human endeavour in the rapidly-evolving landscape of AI.
The Big Picture: If you’re looking to get into the AI industry while it’s hot, learning how to write the right prompts will end up being one of the most employable skills that a young professional can have. It’s fast and affordable and along the way, you gain massive experience in using the tools that are being integrated into nearly every industry today.
(7 min read) (Source: The Guardian)
TLDR: Artificial intelligence (AI) and technologies like ChatGPT have fueled concerns that they will lead to job losses for millions of workers in a variety of industries including advertising, journalism, and computer programming. While AI is likely to eliminate some jobs, many experts argue that it will also create new jobs and enhance existing ones, offsetting much of the job loss.
However, the big question is what type of jobs will be created. White-collar jobs in office-based work, customer service, and sales are expected to have the highest rate of automation adoption and job displacement. Some experts suggest that middle-level, white-collar jobs may be affected more than lower-paying, physically intensive jobs.
On the other hand, AI is expected to create jobs for engineers and certain types of managers. The concern is that AI will reduce the value of some skill sets and move workers into lower-paying jobs. The need for a robust retraining system to prepare workers for new jobs may also become necessary.
The Big Picture: The rise of automation and globalization has led to increased income inequality, as a small group of individuals at the top reap the benefits while workers at the bottom face job loss and wage stagnation. The potential for increased income inequality means that traditional work structures, such as the 5-day workweek, will have to be reevaluated.
Exploring options such as Universal Basic Income and fair redistribution policies to address income disparities could serve as a good starting point in addressing these issues before they become too severe.
(7 min read) (Source: ZDNET)
TLDR: Cerebras and the National Energy Technology Laboratory have found a way to speed up scientific calculations that can now simulate extreme weather conditions in real-time by creating digital twins of real life weather conditions. This was made possible by using a special software package designed for their AI computer chip (the largest computer chip in the world), the Wafer Scale Engine.
Running this on a regular computer would be much slower and not allow for real-time simulations. The results of this study will allow scientists to make predictions and see the effects of different interventions in real-time, helping to improve important projects like carbon sequestration and blue hydrogen production, which are crucial in mitigating climate change and ensuring a secure energy future.
Here is a demonstration of the simulation in action:
The Big Picture: AI is helping scientists perform experiments and simulations that generally require massive processing power at a faster rate than ever before. They speed up important projects that affect time-sensitive issues like climate change which are rapidly becoming worse as the global population grows along with the industries that support it.
💰 Funding Alerts
Acceldata, the leader in data observability, raised $50 million in a Series C funding round led by March Capital and joined by Sanabil Investments, Industry Ventures, and Insight Partners, to speed up innovation and expand their Data Observability Platform's go-to-market efforts.
Canoe Intelligence, a fintech company offering alternative investment intelligence, has completed a $25 million Series B funding round led by F-Prime Capital with participation from Eight Roads Ventures and other investors. The funds will be used to expand into Europe, enhance product offerings, develop new data products, and grow the team.
Entropik, a provider of AI-powered market research platforms, raised $25 million in Series B funding led by Bessemer Venture Partners and SIG Venture Capital. They plan to use the investment to speed up product building in India and accelerate expansion globally to the US, Europe, and Asia.
Avicenna.AI is a provider of healthcare solutions that use AI and deep learning to identify, detect and quantify life-threatening pathologies from CT medical images. In a Series A funding round, they brought their total investment to $10 million. The funds will be used to scale up deployment and diversify the company’s offering into new areas of medicine.
Macro, a startup that created an AI-powered digital document productivity tool, raised $9.3 million in seed funding led by Andreessen Horowitz and others. The investment will aid in their mission to enhance the user experience for digital document workers.
🦾 Trending Tools
🚨 COOL TOOL 🚨 Magic Canvas is a generative AI-based tool that can create physical products from written descriptions in just minutes. Using a latent diffusion neural network, Magic Canvas can turn simple descriptions, such as "a banana riding an elephant” into various physical creations using subtractive technology to cut and engrave products from durable and beautiful materials like wood, leather, acrylic, and paper.
Scalenut, an all-in-one marketing platform, utilizes AI technology to automate the content creation process, saving 90% of the time involved and increasing organic traffic by 10 times. From keyword planning to AI writing, content optimization, and publishing, the app streamlines the entire content lifecycle.
Venngage created an AI card generation tool to generate personalized and unique Valentine card designs.
🌎 Popular Content
1. ICYMI: Here is a replay of today’s announcement from Google. They turned the comments off… (YouTube)
2. ICYMI Part 2: Here is a replay of yesterday’s announcement from Microsoft. They left their comments on… (YouTube)
3. David Guetta is already cooking up some heat with AI but he's not releasing it 🙁 (Instagram)
4. It just keeps getting worse for Google 🤦 Google Bard claimed in a demo posted on Google’s Twitter page that the James Webb Space Telescope was the first to take a picture of a planet outside our solar system. Check out the tweet for the full thread. (Twitter)
Speaking as someone who imaged an exoplanet 14 years before JWST was launched, it feels like you should find a better example?
— Bruce Macintosh (@bmac_astro)
1:48 AM • Feb 8, 2023
5. A Reddit user asked ChatGPT to write instructions for a generative AI on how to draw DaVinci’s Vitruvian Man and then ran the instructions through Midjourney. The results are amazing. (Reddit)
6. Reddit user pro_raze made this handy list of the AI progress that’s been made in the first week of February. Check out the original image to see it in full detail. (Reddit)
👀 More Reading
Five Predictions for the Future of Learning in the Age of AI (Andreessen Horowitz)
‘There is no standard’: investigation finds AI algorithms objectify women’s bodies (The Guardian)
Zero10 wants to crack open 3D and AR fashion design for the masses (Vogue Business)
📚 Academic Papers
Three Strikes and you are out!: The impacts of multiple human–robot trust violations and repairs on robot trustworthiness (Science Direct)
This study explored the impact of four trust repair strategies (apologies, denials, explanations, and promises) between a human and a robot co-worker on overall trustworthiness and its sub-dimensions (ability, benevolence, and integrity) after repeated trust violations by a robot co-worker. The results showed that none of the strategies fully repaired trust after repeated violations and that denials were consistently the least effective. Apologies, explanations, and promises appeared to function similarly.
📅 Upcoming Events
The World AI Cannes Festival (TODAY February 9-11, 2023. Cannes, France + Virtual), where decision-makers and AI innovators meet, where the most promising innovations and technologies get into the spotlight, where those who are currently building the world’s most game-changing AI strategies and use-cases will be on stages.
The Gen AI Conference (February 14, 2023. San Francisco, CA) hosted by Jasper AI, is the first-ever generative AI conference. Attendees can learn about the many recent developments in the field of AI from experts and network with like-minded individuals in AI, business, and marketing. Online registration will close on February 13th at 11:59 PM Pacific Time.
RE•WORK AI Summit West, Deep Learning Summit (February 15-16, 2023. San Francisco, CA), a chance to hear the latest technology advancements, practical examples of how to apply AI to solve challenges in cross-industry settings, business and society, and delve deeper into the work of leading AI experts in a series of presentations, panel discussions, interviews and fireside chats.
Data Science Salon Austin (February 21-22, 2023. Austin, TX + Virtual) is a two-day 500-person conference focused on AI and machine learning applications in the enterprise. The intimate event curates data science sessions to bring industry leaders and specialists face-to-face to educate each other on innovative new solutions in artificial intelligence, machine learning, predictive analytics and acceptance around best practices.
💼 Jobs
That's all for todays email! If you want more please follow us at the social channels linked below, or check out our website!
How'd you like today's email? |
Share our newsletter: If you like our work please share/forward this email with your friends, colleagues, and family. It's the best way to support us!
If this email was forwarded to you please sign up here to continue receiving them.
Want your content, product, jobs, or event featured in our newsletter? Reply to this email with the details, and our team will reach out to you.
Do you use AI for work? Tell us how, and you could be featured in our newsletter!
Check out our website for more resources, including a list of AI investors, products, events, and twitter follows.
For an archive of all our posts, click here.
We'd love to hear from you! You can always leave us comments or feedback by replying to this email!