What happened: Education technology firm Chegg Inc. saw its shares tumble 48% on Tuesday after its CEO cautioned that ChatGPT was cutting into growth. Details: The Santa Clara, California-based firm offers subscription-based services to help students with their coursework. During an investor call Monday, CEO Dan Rosensweig said ChatGPT is impacting Chegg's customer-growth rates after seeing a spike in students' ChatGPT interest since March. While not a "substantial" impact, Rosenberg said users who would typically pay for Chegg near midterms or finals are outsourcing to freely available ChatGPT. Other impacts: Tuesday's selloff also impacted shares of Duolingo and Pearson. Chegg has also pulled its annual revenue forecast from February, citing uncertainty around ChatGPT's impact on online-learning platforms. Still: Chegg plans to incorporate AI into its own services. Last month it launched CheggMate, a GPT 4-powered AI study guide for students. 
| |
Samsung Electronics has temporarily barred employees from using generative AI tools after an engineer unintentionally uploaded internal source code to ChatGPT. The company is concerned that data shared with AI chatbots and other tools may be stored on external servers operated by Google, Microsoft, and others, with no straightforward way to access or delete it. More: - Samsung is worried sensitive information shared on ChatGPT could be disclosed to other users and/or accessed by companies such as OpenAI.
- Its ban applies to generative AI services on company-owned devices as well as any other devices running on internal networks.
- The company said its ban is temporary while it works to "create a secure environment" for generative AI tools.
- In an internal survey, Samsung found that 65% of respondents believe that AI chatbots and other services are a security risk.
- In the case of ChatGPT, inputs on the platform's general web interface are used to train the company's AI models by default, meaning future systems could accidentally share a company’s private data.
CNBC 
| |
 A message from TRUERA LLM Testing and Monitoring – Join the Hands-On Workshop How can you ensure that your large language model is honest, harmless, and helpful? Join TruEra on May 11th and 18th as their President and Chief Scientist, Anupam Datta, provides a hands-on overview of how to analyze and improve the performance of your LLMs. During the two 60-minute online sessions, he will cover the following: - Quick introduction to LLMs - what are they and their history?
- Risks and challenges of LLMs: is your LLM honest, harmless, and helpful?
- Defining feedback functions to model LLM quality and performance
- How should you test and monitor LLMs?
- Live walk-throughs of how to identify, diagnose, and debug model issues in development and production
- Q&A with Professor Datta
Dates: 2-session series on May 11th and 18th Where: Virtual Event - Join from Anywhere Who should attend: Data scientists, machine learning engineers, MLOps managers, data analysts, statisticians, data and analytics managers Join for free
| |
Microsoft's Azure cloud division plans to offer a specialized version of ChatGPT that will operate on dedicated rather than shared cloud servers. The version will store data separately from that of other users, addressing some concerns about data and information leaks. However, it will come at a higher cost, potentially up to 10 times more expensive than the standard version of the chatbot, according to reports. More: - The specialized version of ChatGPT is aimed at attracting businesses, such as banks and healthcare institutions, that are concerned about employees sharing sensitive proprietary information with the chatbot.
- JPMorgan Chase & Co., Bank of America, Citigroup, Goldman Sachs Group, and other banks have all restricted employee access to ChatGPT.
- According to The Information, OpenAI recently forged a deal with Morgan Stanley for a private GPT-4 service aimed at its financial advisers.
- Its wealth management division will have dedicated servers running GPT-4.
- Microsoft first announced the Azure OpenAI Service in January and later made ChatGPT generally available through the service.
| |
U.S. Rep. Yvette Clarke (D-N.Y.) has proposed a bill that would require political campaigns and groups to reveal their use of AI-generated imagery in political ads. Clarke said her bill is a direct response to the Republican National Committee's recent AI-produced ad depicting dystopian images and scenarios that it suggests could happen if President Biden were re-elected. More: - Clarke's bill aims to amend federal campaign finance laws to require political ads to disclose any AI-created content, as well as compel the Federal Election Commission (FEC) to create rules to enforce the disclosure.
- If passed into law, the requirement would take effect on Jan. 1, 2024.
- Clarke noted in a news release that the 2024 elections will be the first time in U.S. history that generative AI is used in political advertising. The technology can "manipulate and deceive people on a large scale," she said.
- While the RNC tagged its AI footage with a disclaimer saying "built entirely with AI imagery," Clarke argued that other political groups might not follow suit.
- "I think there are really important uses of AI, but there have to be some rules to the road so that the American people are not deceived or put into harm's way," she said.
THE WASHINGTON POST
| |
 A message from VANTA Swamped with spreadsheets as you work to stay compliant with privacy and security regulations? Then you need Vanta, the market-leading automated compliance platform. Trusted by 5,000+ global customers, Vanta helps you quickly and easily get compliant, manage risk, and show off your security — so your business can focus on building and selling. Vanta’s platform: -
Automates up to 90% of compliance for SOC 2, ISO 27001, GDPR, HIPAA, and more, saving you hundreds of hours of manual work -
Gives you one place to centralize and scale your security program and build trust with customers and partners -
Helps your business scale and thrive while reducing the need for countless spreadsheets and endless email threads Even better? Vanta customers report saving up to 85% of compliance costs. And now, you can take a self-guided tour of Vanta’s product to see how it all works. See the Tour
| |
Sound Ventures, a venture-capital firm co-founded by Ashton Kutcher and Guy Oseary, has closed on a $240M oversubscribed fund to invest in AI companies. The Sound Ventures AI Fund currently has stakes in OpenAI, StabilityAI, and Anthropic, which all offer generative AI systems. More: - The fund will focus on companies at a "foundation model layer," meaning businesses offering large-scale pre-trained AI models that can be adapted by others across a wide range of tasks.
- In a news release, Kutcher called AI "potentially the most significant technology" since the internet. Foundation model layer firms are "defining the category" and "have the power to transform businesses and everyday life," he said.
- Sound Ventures, led by Kutcher, Oseary, and Effie Epstein as general partners, currently manages over $1B in assets across funds, which have previously backed Spotify, Pinterest, Airbnb, Brex, GitLab, and others.
MARKETS INSIDER
| |
New job postings in AI: - Director, Machine Learning Engineering (Enterprise Feature Platform)
- Capital One | New York, New York and McLean, Virginia
- Data and Applied Scientist II
- Microsoft | Redmond, Washington
- R&D Computer Science — Scientific Machine Learning
- Sandia National Laboratories | Albuquerque, New Mexico
- Senior Staff, Solutions R&D — Artificial Intelligence EDA Solutions
- Synopsys | Mountain View, California
- Research Scientist — Salesforce AI Research
- Salesforce | Palo Alto, California
- Senior Machine Learning Engineer, Ads Targeting
- Senior Research Scientist — Conversational AI
- Nvidia | Santa Clara, California
| |
QUICK HITS - The company behind these surgical robots is planning to list on Nasdaq. Last day to fund your investment is on 5/10.*
- Amnesty International has stood by its decision to depict human rights abuses using AI-generated images. The human rights organization created and shared three fake images of police brutality against protestors in Colombia.
- Italy has lifted a ban on OpenAI's ChatGPT chatbot after the U.S.-based AI company addressed privacy concerns raised by Italy's data protection regulator.
- LinkedIn premium subscribers can now access a generative AI feature that automatically drafts brief messages, similar to cover letters, that job seekers can send to hiring managers.
- Demo Divvy, get an Ember heated mug. Hot coffee and quick and easy expense management — get both when you demo Divvy.*
*This is a sponsored post.
| |
Upcoming events: *This is a sponsored listing.
| |
 |  | Beth is a writer and editor at Inside.com covering artificial intelligence and daily technology news. She has written the Inside AI newsletter since 2019. You can reach her at Beth.Duckett@inside.com. |   | | Editor | Aaron Crutchfield is based in the high desert of California. Over the last two decades, he has spent time writing and editing at various local newspapers and defense contractors in California. When he's not working, he can often be found looking at the latest memes with his kids or working on his 1962 and 1972 Fords. | |
Join us on May 11th and 18th for a hands-on overview of how to analyze and improve the performance of your LLMs. | |
| |
Comments
Post a Comment