| |
| Welcome to the latest AI newsletter! Here are today's top stories: - Databricks secures $500M in funding round.
- California bill would require testing of advanced AI models.
- DHS introduces guidelines for its AI use.
If you find this newsletter useful, please consider sharing it with your friends and colleagues. Thank you! Beth
p/beth-duckett | |
| 1 | | Databricks, an AI and data analytics software company, has secured over $500M in funding, valuing the company at $43B. The latest funding round has made Databricks the world's eighth most valuable private company, per Crunchbase data. More: - The Series I round was led by T. Rowe Price, with participation from Nvidia, Capital One Financial, and others.
- In August 2021, Databricks raised $1.6B at a $38B valuation. The company's increased valuation in the current market is linked to its strong financial performance, with sales up 50% YoY and a revenue run rate of over $1.5B in Q2 2023.
- The company, founded in 2013, has gained recognition for its cloud-based data processing software.
- It recently acquired generative AI startup MosaicML for $1.3B, further expanding its AI capabilities.
- In June, Databricks introduced LakehouseIQ, an AI assistant that understands natural language queries.
|     | |
| 2 | | California Senator Scott Wiener has proposed a state bill that seeks to regulate powerful generative AI models and require labs to test advanced models for safety risks. California and Maryland have already taken the lead in crafting AI regulations, passing nearly 50 AI-related bills from 2016 through 2022. 
Details: Wiener's newly unveiled SB 294, the Safety in Artificial Intelligence Act, seeks to create a state agency or delegate the role to an existing one to oversee responsible AI development and hold developers accountable. It would mandate transparency standards for "frontier" model systems, or large language models, when they exceed a certain computing power. The "intent bill" also seeks to create CalCompute, a cloud-based compute cluster for AI researchers and smaller developers in the state's public university system. What the numbers show: From 2016 through 2022, California passed a total of six AI-related laws, second to only Maryland in the number of bills signed into law. So far this year, 25 states, along with Puerto Rico and D.C., have introduced AI bills, including California's Assembly Bill 459, which would provide greater protections for actors from AI voice and body cloning. Related: Earlier this month, California Gov. Gavin Newsom signed an executive order on AI, directing state agencies to explore the uses and risks of AI and generative AI technologies. While other governments are proposing stricter AI regulations, Newsom's order takes a lighter approach, emphasizing California's role in fostering AI innovation. The executive order is seen as a first step in regulating AI in California, where 35 of the top 50 AI companies are located. Q: Do you think that tighter AI regulations are needed to promote responsible AI development, or could they hinder innovation in California? Join the conversation here. | | | |
| |
|
| 3 | | The U.S. Department of Homeland Security (DHS) introduced AI guidelines allowing Americans to opt out of facial recognition technology in air travel check-ins and other situations. The policy also requires manual human review of AI-detected facial recognition matches to ensure accuracy. More: - The new principles demand that DHS refrain from collecting, using, or sharing data used in AI activities.
- The rules prevent DHS from creating AI systems that use factors like race, gender, or ethnicity to make or support decisions.
- A directive calls for testing of all facial recognition and face capture technology to eliminate unintended bias.
- The guidelines were developed by the new DHS AI task force. Additionally, DHS has appointed Eric Hysen, the department's chief information officer, as its first chief AI officer.
Zoom out: - DHS uses AI for missions like combating fentanyl trafficking and countering child exploitation.
- The agency has extensively employed AI technology on the southern border, deploying over 200 surveillance cameras from Anduril that use AI to detect and flag human crossings.
- Customs and Border Patrol agents have also used advanced machine learning models to flag "suspicious" looking vehicles.
| | | |
| 4 | | Generate Biomedicines, an AI-driven drug discovery startup, secured $273M in a Series C funding round, with backing from investors including Amgen and Nvidia's VC arm. Generate, founded by Moderna incubator Flagship Pioneering, leverages machine learning to identify cell therapies, antibodies, peptides, and other medicines. More: - Generate utilizes AI to replace trial-and-error in drug discovery, crafting novel protein-based therapeutics.
- The company says its Generate Platform speeds up drug target and therapeutic identification and validation through a continuous loop of generation, construction, measurement, and learning.
- The biotech initiated its first COVID-19 monoclonal antibody clinical trial in July and is progressing toward another trial for an asthma treatment.
- It currently has 15 drug candidates in development and aims to introduce 10 new ones each year.
- Since 2020, Generate has secured ~$700M in equity financing.
| | | |
| A message from our sponsor, Sidebar . |  | | Ready to take your career to the next level? Make your transition successful by leveraging a Personal Board of Directors. A trusted peer group, with battle-tested perspective, and proven playbooks, to have real, tactical discussions with to propel you forward. With Sidebar, senior leaders are matched with a small group of highly vetted, private, and supportive peers to lean on for unbiased opinions, diverse perspectives, and raw feedback. - “Providing and receiving support from others who play a similar role to you is one of the best ways to grow your capabilities and succeed.” - Vice President, Roku
- “You’re the average of the people you keep closest; Sidebar helps you raise that bar.” - Global Director, Reddit
- “Tap into a new level of coaching support, professional development, and peer-driven accountability.” - Senior Director, Credit Karma
Why spend a decade finding your people – join the growing waitlist of over 5,000 top senior leaders, and apply to become a founding member today. Join the Waitlist | |
|
| 5 | | NASA plans to use AI and machine learning to investigate and search for unidentified aerial phenomena (UAP). The space agency released a 33-page report summarizing findings from an independent yearlong study commissioned by NASA on UAPs. NASA plans to follow some of the report's recommendations, such as using AI and machine learning to study UAP. More: - The study found no evidence of extraterrestrial origin but admitted a lack of explanation about some UAP sightings.
- It highlights that AI and machine learning are crucial for identifying rare events in large datasets but stresses that the quality of UAP data is a limiting factor.
- The study team suggests that neural network techniques used in other fields can be applied to UAP research, provided that data quality is enhanced and standardized.
- The team suggests that NASA leverage open-source resources, its technological expertise, data analysis techniques, partnerships, and Earth-observing assets to improve its UAP dataset.
- NASA's UAP research director will employ AI and machine learning for data gathering and analysis.
| | | |
| 6 | | Quick Hits: - Before launching your next influencer campaign, compare the latest cost metrics to your budget. Get the benchmarks in this blog.*
- Pixis, a provider of codeless AI infrastructure for performance marketing, secured $85M in series C1 funding led by Touring Capital, bringing its total capital raised to $209M.
- Microsoft introduced EvoDiff, an open-source AI system and framework that generates proteins given a sequence. Use cases include drug delivery, therapeutics, and industrial reactions.
- AI startup Anthropic has partnered with Boston Consulting Group (BCG) to provide BCG clients with direct access to its Claude 2 large language model and AI technology.
- Hyperstack: Scale AI projects with NVIDIA GPUs. Enjoy per-minute billing. Secure your H100s & A100s today!*
*This is sponsored content. | | | |
| Upcoming Events | | SEP
21 | | Join us for pitches, drinks and networking at HanaHaus in Palo Alto on September 21 at 6pm.* | | | | | | | OCT
16 | | Meet Our Fund 5, Building and Investing in AI | | | | | | * This is a sponsored event | | | |
Term of the Day Cash burn: Cash burn is defined as the amount of cash a company utilizes to meet its expenses in a specific period. Question of the Week Do you share details about your salary with your coworkers? Join the conversation |
INSIDE AI LEADERBOARD (7 DAYS) |
 | AI and technology writer | | Beth is the writer of Inside's AI and Tech newsletters, with a career spanning journalism and editing since 2007. She has written for publications including The Arizona Republic, USA Today, and others, and authored a book on the solar industry in 2016. Reach out to her at Beth.Duckett@yahoo.com. | | This newsletter was edited by Aaron Crutchfield | |
| |
| |
|
 | Sidebar – Catalyze your career with a Personal Board of Directors. | |
|
| |
Comments
Post a Comment