- Work With AI
- Posts
- Researchers from Google and Stanford Think AI Is Entering an Era of Corporate Control
Researchers from Google and Stanford Think AI Is Entering an Era of Corporate Control
April 4, 2023
⚡ Today’s Highlights
📰 News: AI is being taken over by corporations and Samsung workers leaked confidential information to ChatGPT
💰 Funding: Quantexa, Fourthline, Redefine, and NuMind
🦾 Tools: Free AI code translation and personalized AI for PR
📅 Events: The Data Dividend and Healthcare NLP Summit start today!
📰 Today's Top Stories
(6 min read) (Source: The Verge)
TLDR: The 2023 AI Index report, compiled by researchers from Stanford University and AI companies such as Google, Anthropic, and Hugging Face, highlights the increasing dominance of industry players over academia and government in deploying and safeguarding AI applications.
Industry players have taken over from academia in developing state-of-the-art AI systems due to the increasingly large resource demands required to create such applications, firmly shifting the balance of power toward corporate players.
The report notes that as AI applications become more widely deployed, the number of incidents of ethical misuse has increased, including fatalities involving self-driving software and the creation of nonconsensual deepfake nudes, leading to growing calls for AI regulation.
The report covers how private investment in AI decreased for the first time in a decade, the environmental costs of training big AI models, the potential for AI to help reduce emissions, and regional differences in attitudes towards AI, with Chinese people more hopeful about AI than Americans.
(3 min read) (Source: Economist Korea)
TLDR: It didn’t take long. 20 days since Samsung Electronics in Korea allowed Samsung Semiconductor engineers to use ChatGPT, 3 cases of misuse that have resulted in the leak of confidential and proprietary information have been leaked.
These engineers in question uploaded confidential information to fix errors in code, requested optimization of test patterns, and converted meeting recordings into presentations in ChatGPT.
Samsung has since warned its employees of the dangers of using ChatGPT due to the risks of exposing sensitive information to third parties, while preparing protective measures and considering blocking access to ChatGPT on the company network in case of another incident.
(7 min read) (Source: VentureBeat)
TLDR: Not many people have talked about their experiences with Claude since its March 14th release. But Anthropic apparently offered early access to its AI chatbot interface, Claude, to select attendees at a Hugging Face-hosted Open Source AI meetup.
This unique marketing strategy allowed industry experts to get early glimpses of Claude’s Constitutional AI, a groundbreaking technique for aligning AI systems with human intentions.
Anthropic aims to use Constitutional AI to develop more helpful, safer, and more robust AI systems that adhere to ethical and behavioural principles, and Claude has undergone extensive safety evaluations.
💰 Funding Alerts
Quantexa has raised $129 million in a Series E funding round led by Singapore’s sovereign wealth fund GIC, with participation from Warburg Pincus, Dawn Capital, British Patient Capital, Evolution Equity Partners, HSBC, BNY Mellon, ABN AMRO, and AlbionVC. Quantexa uses AI and other tools to provide anti-fraud and anti-crime solutions to businesses, governments and other major organizations. The funding will be used to invest in new technology and make acquisitions to fuel growth. The round values the company at $1.8 billion, up from its previous valuation of $800 million to $900 million in 2021.
Fourthline announced that it had raised €50 million in funding led by Finch Capital (Radboud Vlaar, Managing Partner) to transform financial compliance for banks and fintech using their AI-powered know-your-customer (KYC) and anti-money laundering solutions. The Fourthline platform provides a suite of proprietary tech products that adhere to local requirements and offer solutions for the entire lifecycle of compliance, from onboarding to continuous KYC. The funds will be used to further develop their platform and establish it as the leading technology solution for compliance in Europe.
Redefine, an AI-powered Continuous Integration optimization platform, raised $8.5 million in a Seed funding round led by Gillot Capital, with participation from industry leaders such as Nitzan Kapira and Ran Rabenzaft (founders of Epsagon) and Efrat Rapoport (director of Salesforce Israel). Redefine’s solutions allow users to accelerate development, reduce run times, and minimize cloud expenses while improving code quality. The company plans to use the funds to grow and accelerate the development of its tools.
NuMind, a natural language processing software startup, has raised $3 million in seed funding from investors including Flybridge, Carya Ventures, and Y Combinator. The company uses large language models and interactive AI development to help customers create machine learning models for NLP tasks such as sentiment analysis and chatbot creation. NuMind's solution is available as a monthly subscription, and the company plans to incorporate generation tasks like summarization and translation later on.
🦾 Trending Tools
This free AI Code Translator uses GPT-3.5/4 to provide code translation between programming languages. You can also generate code from natural language instruction and get natural language explanations for code. All you need is an OpenAI API Key.
Write Release, created by PR agency Gregory FCA, lets you create high-quality, non-cliché press releases in minutes. Answer a few questions to generate your first draft, pick your tone, and choose a headline to get started. Free and powered by AI, it's the PR tool you never knew you needed.
Butternut.ai is an AI-powered website builder that lets you generate stunning websites in seconds, without any coding required. Enter your business name and keywords, and get a fully-functional website with images and copy in under 20 seconds.
DiffusionArt is a free, web-based, anonymous alternative to MidJourney for creating unlimited artwork without having to share it or log in to Discord. It's accessible from any device, has a built-in advanced prompt generator and tuner, and offers tools like an upscaler, image variation generator, and image-to-text generator.
👀 More Reading
Silicon Valley school embraces a new AI tool as others ban ChatGPT (Washington Post)
Does AI Dream of Electric Sheep? (Wall Street Journal)
Can the Government Get a Handle on Artificial Intelligence? (The Atlantic)
Hugging Face hosts ‘Woodstock of AI,’ emerges as leading voice for open-source AI development (VentureBeat)
🌎 Fresh Content
1. Reid Hoffman of Linkedin and Greylock sat down to discuss the current state of AI with OpenAI CEO Sam Altman on the Greymatter podcast. Hoffman’s recent book, “Impromptu”, was co-written by GPT-4. They discuss the pros and cons of AI advancements and their immediate impacts and why safety is key to AI development above all. (Greymatter Podcast)
2. Hank Green of vlogbrothers had some amazing insights on how AI is advancing and how it might affect our society and explains it using the wonderful analogy of rats and mongooses in Hawaii. (YouTube)
3. Marques Brownlee had a great theory about the way that generative AI is evolving and the point at which it’ll get really scary. The example that he provides is that of AllttA’s new song, “Savages”, which features a verse from Jay-Z that is completely AI-generated. The song was very popular on social media and it will be quite interesting to see how it does considering that another song made using an AI-generated Eminem verse was taken down because of a DMCA strike from Universal Music Publishing Group. (YouTube)
Universal Music Group just decided to copyright strike my AI Eminem Cat Rap video. Usually they just claim and monetize videos like this but they really wanted AI Eminem to be taken down for some reason.
— Dr Grandayy (@grandayy)
2:09 PM • Apr 2, 2023
🧠 Resources
We've developed these prompts to help you explore how you can use AI frameworks to improve your marketing campaigns and create a sense of obligation, emphasize internal qualities, and shape your customer's perceptions about your product or service. As you begin to implement these AI frameworks into your marketing campaigns, don't forget to keep track of the results. We encourage you to share your feedback with us and let us know how these prompts worked for you.
Develop a marketing campaign strategy that utilizes the concept of 'Reciprocity Bias' to persuade [ideal customer profile] to try [product/service]. In addition to highlighting the benefits of the product/service, include value-adds or bonuses, and encourage reciprocity by asking for a favor or action in return.
Construct a marketing campaign that utilizes the 'Attribution Bias' framework to emphasize the internal qualities of [product/service] and how it can help [ideal customer profile] achieve their goals. By attributing successes or failures to internal factors, create a persuasive narrative that showcases the strengths of [product/service].
Craft a marketing campaign that employs the 'Anchoring Bias' framework to influence the perceptions of [ideal customer profile] about [product/service]. By strategically highlighting the most important or relevant information first, create an anchor that shapes their decisions and helps them see the value of [product/service].
📅 Upcoming Events
Introducing The data dividend: harnessing the power of people, processes and technology to unlock value from data (April 4, 2023 / Boston, MA, USA) a forum that brings together leaders in government and industry to discuss how to establish the strategy, culture and skills needed to foster a data-driven enterprise and cultivate new ways of thinking and working. They’ll explore questions like:
What is top of mind for data and analytics leaders in 2023? What key challenges are they facing?
How is the regulation of data and AI evolving, and how will it affect your industry and organisation to bring new opportunities and challenges?
Can we trust decisions and recommendations made by AI? How can tech leaders mitigate risk, bias and harm?
Healthcare NLP Summit (April 4-5, 2023 / Virtual) is a two day event consisting of immersive, industry-focused content including over 30 technical sessions, focusing on NLP applications in healthcare and life sciences. Join live Q&A sessions with the speakers, connect with others through networking features, and access all content on-demand after the event. The second week will feature beginner to advanced live training workshops and certifications. Learn, share, and apply best practices for putting AI to good use! Register for this free online event today!
💼 Jobs
Phantom AI: Senior Computer Vision Engineer - Deep Learning Focus
Mountain View, CA, USA
$170,000 - $215,000 USD
Infinitus: Insurance verification Specialist / Artificial Intelligence Trainer (Arizona)
San Francisco, CA, USA (Remote)
$17/hour USD
Columbia, MD, USA
$100,000 - $145,000 USD
Tubi: Senior Tech Lead, Machine Learning
San Francisco, CA, USA (Remote)
$198,000 - $252,000 USD
Lightmatter: Machine Learning Research Engineer
Mountain View, CA, USA
$200,000 - $230,000 USD
That's all for today's 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!