• AI KATANA
  • Posts
  • 82% Americans don't trust tech executives to regulate AI

82% Americans don't trust tech executives to regulate AI

Also: GenAI 50: The most promising generative artificial intelligence startups of 2023

Author

AI KATANA
August 11, 2023

Welcome!

With public sentiment leaning towards government regulation of AI, the Pentagon's newly established AI task force, and San Francisco poised to be an AI hub, it's evident that AI continues to be at the forefront of tech innovation and governance. This week, we also highlight groundbreaking updates from Stability AI and Anthropic, two pioneers in the realm of AI tools. In the venture capital segment, AI-driven platforms Middleware and Symmetry Systems have secured notable funding, emphasizing the industry's robust growth and dynamism. Dive in for a comprehensive review of the week's most pertinent AI updates.

Sliced:

  • 🇺🇸 82% Americans don't trust tech executives to regulate AI

  • ⚠️ The Pentagon just launched a generative AI task force

  • 🤖 GenAI 50: The most promising generative artificial intelligence startups of 2023

  • 🌉 AI boom could fuel San Francisco's comeback

A recent poll by the Artificial Intelligence Policy Institute reveals that a significant majority of American voters across political affiliations are concerned about the risks associated with AI and are in favor of federal regulation to mitigate these risks. The poll, shared exclusively with Axios and conducted by YouGov, shows that 82% of respondents do not trust tech executives to self-regulate AI, while 56% support the idea of a dedicated federal agency for AI regulation. These numbers reflect a growing distrust in tech companies and indicate a broader global trend toward the desire for AI regulation, as evidenced by similar sentiments in surveys from other countries. However, there remains skepticism about the capability of governments, especially in democracies, to appropriately regulate AI.

The Department of Defense (DoD) has launched a generative AI task force named Task Force Lima, reflecting its dedication to responsibly leveraging AI technology. Directed by Deputy Secretary of Defense Dr. Kathleen Hicks and led by the Chief Digital and Artificial Intelligence Office (CDAO), the task force will integrate and analyze generative AI tools, including large language models, across the DoD. The initiative aims to maintain the DoD's technological leadership while ensuring national security and addressing potential risks. Partnering with various government agencies, the task force emphasizes responsible adoption of AI to enhance areas like warfighting, health, and policy. The CDAO, operational since June 2022, oversees the integration of AI within the DoD.

CB Insights has revealed its inaugural GenAI 50 list, spotlighting the 50 most promising private generative AI companies worldwide. This year's distinguished winners comprise companies focusing on diverse areas such as foundation models, code generation, and AI assistants. Of the entire cohort, 16 are honing in on industry-specific applications, including character animation in media and entertainment or drug discovery in healthcare. Meanwhile, 20 vendors are crafting cross-industry solutions like AI assistants, human-machine interfaces, and code generation tools. The remaining 14 companies are innovating tools such as vector database technologies and foundation models to bolster AI development.

San Francisco's downtown, which has been struggling, may see a resurgence due to the boom in AI companies. Office space searches in the city have surged, with a quarter attributed to AI businesses, indicating a level of demand not witnessed since the introduction of COVID vaccines in 2021. San Francisco is becoming a hub for AI development, hosting 20 of the top 50 AI companies as listed by Forbes. As per JLL, a commercial real estate company, ten AI firms are currently scouting for nearly 700,000 to 800,000 square feet of office space in the city. Despite the optimism, there's still a vast amount of vacant office space in the city, and smaller AI startups don't yet have the workforce to occupy large downtown skyscraper floors. However, some believe the city is transitioning to become a novel kind of knowledge center, with other cities like Boston, Seattle, and Washington, D.C., expected to follow suit.

🛠️ AI tools updates

Stability AI has launched its premier Japanese language model, Japanese StableLM Alpha, which boasts 7 billion parameters and stands as the top-performing public Japanese LM according to benchmarks. Two versions of the model have been released: Japanese StableLM Base Alpha 7B, trained primarily with Japanese and English text from various sources, including Stability AI Japan and EleutherAI Polyglot project datasets, and Japanese StableLM Instruct Alpha 7B, designed to follow user instructions with improved performance scores in various tasks. Both models are available on Hugging Face Hub for inference and further training, with Japanese StableLM Instruct Alpha 7B attaining a score of 54.71, outpacing other Japanese models.

Anthropic has launched its newest model, Claude Instant 1.2, offering faster, more affordable, and efficient services for tasks like dialogue, text analysis, and document comprehension. Drawing from the strengths of Claude 2, this update showcases major advancements in areas such as math, coding, reasoning, and safety. Notably, Instant 1.2 has shown improved scores on benchmarks like Codex and GSM8K compared to its predecessor, Claude Instant 1.1. The model also offers enhanced safety features, reducing hallucinations and increasing resistance to jailbreaks. Businesses and developers interested in Claude Instant 1.2 can access it via API, with detailed pricing and partnership opportunities available.

💵 Venture Capital updates

Middleware, an AI-powered cloud observability platform, announced a $6.5 million seed funding round to enhance cloud observability capabilities. Led by 8VC, participants included Fin Capital, Vercel CEO Guillermo Rauch, Tokyo Black, and angel investors like Decent Capital, Begin Capital, Beat Venture, and Gokul Rajaram. The funds will be directed towards team expansion, new feature development, and customer base growth.

Symmetry Systems, an AI-driven data security firm of the next generation, has secured $17.7 million in a funding round. This round saw continued involvement from ForgePoint Capital and Prefix Capital, alongside new entrants W11 Capital Management and TSG (The Syndicate Group), a strategic investor with a focus on channels.

🫡 Meme of the day

⭐️ Generative AI image of the day