Writer secures $100M to Advance enterprise-focused generative AI

Pivot 5: 5 stories. 5 minutes a day. 5 days a week.

1. Writer secures $100M to Advance enterprise-focused generative AI

Writer, a startup specializing in generative AI for businesses, has successfully raised $100 million in a Series B funding round. Led by ICONIQ Growth and featuring participation from Accenture and Vanguard, the funding boosts Writer's total capital to $126 million and values the company between $500 million and $750 million. The new funds will be used to develop industry-specific text-generating AI models, according to CEO May Habib.

Enterprises often find it challenging to implement generative AI due to the complexities involved in data gathering, cleaning, and workflow construction. Writer aims to alleviate these issues by offering a full-stack generative AI platform that can be hosted in an enterprise virtual private cloud. This makes it easier for businesses to adopt generative AI for various use cases, from internal applications to customer-facing solutions.

Writer faces stiff competition from other generative AI platforms like OpenAI, Jasper, and Cohere. However, it distinguishes itself by training its models on non-copyrighted business writing and offering transparent, cost-effective solutions. These features have attracted a diverse customer base, including Intuit, Uber, and Spotify. Based in San Francisco, Writer claims to have grown its revenues by 10x in the last two years, signaling strong market traction.

Read the full story here 

2. MIT's multi-agent AI system aims to enhance reasoning in language models

MIT's Computer Science and Artificial Intelligence Laboratory has developed a multi-agent AI system designed to improve the factual accuracy and reasoning of language models. The system allows multiple AI agents to discuss and critique each other's responses, leading to a more accurate and reasoned final answer. This approach addresses the problem of inconsistencies and inaccuracies often found in large language models.

The multi-agent system works through an iterative process where each AI agent generates a response to a given question and then incorporates feedback from other agents to refine its answer. This process culminates in a majority-vote-based final output, mimicking the dynamics of a group discussion. The approach can be applied to existing black-box models, offering a straightforward way to improve the consistency and factual accuracy of language model outputs.

While the system has shown promise, especially in mathematical problem-solving, it still faces challenges. These include limitations in processing very long contexts and refining critique abilities. However, the researchers believe that the multi-agent debate format could be a crucial area for future exploration. As AI models continue to evolve, this collaborative approach could pave the way for more reliable and accurate language understanding and applications.

Read the full story here 

3. Google and U.S. Department of Defense develop AI-powered microscope to aid cancer diagnosis

Google and the U.S. Department of Defense have collaborated to create an Augmented Reality Microscope (ARM), designed to assist pathologists in diagnosing cancer. Dr. Nadeem Zafar, a pathologist at the VA hospital in Seattle, recently used the ARM to settle a diagnostic disagreement with a colleague. The AI-powered microscope quickly identified the aggressive part of a tumor, validating Dr. Zafar's assessment.

The ARM is more than just a high-tech microscope; it's a blend of traditional pathology and cutting-edge AI. When a slide is placed under the microscope, the AI outlines the cancerous area, providing a heat map that indicates the severity of the cancer. This technology aims to serve as a second line of defense for pathologists, especially those without easy access to a second opinion. It also addresses the challenges of digital pathology, which can be costly and data-intensive, by offering a less expensive alternative.

The ARM is still in its early stages, but its potential is significant. Currently priced between $90,000 to $100,000, the technology is undergoing rigorous testing and peer review. The Defense Innovation Unit at the Department of Defense is considering scaling the technology and collaborating with regulators. As Dr. Zafar notes, "AI is here, and it's going to keep developing. The point is not to be afraid of these technologies, but to triage them to the best use for our medical and healthcare needs.

Read the full story here

4. Betterleap raises $13M to transform recruitment with AI-driven insights

San Francisco-based startup Betterleap has officially launched with an impressive $13 million in seed funding from leading venture capital firms like a16z and Peakstate Ventures. The company aims to disrupt the recruitment industry by leveraging AI and data analytics. Betterleap boasts the world's largest database of job candidates, with over 1 billion records from various platforms like LinkedIn and GitHub. Its generative AI platform provides recruiters with actionable insights to find the best candidates for their roles.

Betterleap's AI engine, CoPilot, offers unique features such as GPT-4 integration and comprehensive candidate mapping. It allows recruiters to ask critical questions and receive immediate responses, streamlining the decision-making process. The platform also supports cross-sector compatibility, providing different ways to engage candidates depending on their industry. CEO Khaled Hussein emphasizes that Betterleap aims to help recruiters "do more with less," focusing on efficiency and resource optimization.

Diversity and inclusion are also at the forefront of Betterleap's mission. The platform allows recruiters to source diverse candidates using various filters and employs AI to analyze the diversity of the recruitment pipeline. Hussein says the seed funding will be used to grow the team, invest in R&D, and expand the company's go-to-market strategy. As the recruitment industry continues to evolve, Betterleap's approach could set a new standard for efficiency and precision, making it a startup to watch closely.

Read the full story here 

5. AI-generated geometric artwork mesmerizes social media

A Reddit user named "Ugleh" recently posted an AI-generated image of a spiral-shaped medieval village that has taken social media by storm. Created using Stable Diffusion and a guidance technique called ControlNet, the artwork has garnered over 145,000 likes on Twitter and praise from Y-Combinator co-founder Paul Graham, who said, "This was the point where AI-generated art passed the Turing Test for me."

ControlNet, a technology that first appeared in a research paper in February 2023, adds an extra layer of guidance to the Stable Diffusion neural network model. It allows for the incorporation of extracted information from a source image, such as pose detection and edge mapping. This enables the AI to generate artwork that closely replicates the shape or pose of a subject in an image. Ugleh's artwork, guided by the prompt "Medieval village scene with busy streets and castle in the distance," showcases the capabilities of ControlNet in rendering intricate geometric scenes.

While the artwork has received widespread acclaim, it has also faced criticism for its compositional elements. Graphic designer Trent pointed out issues like incorrect shadows and illogical placements of chimneys above windows. Moreover, the artwork's copyright status remains uncertain, as current U.S. policy suggests that AI-generated images may not meet the standards for copyright protection. Despite these challenges, the artwork exemplifies the creative potential of AI, even as it raises questions about the future of AI-generated art.

Read the full story here 

Listen to the daily Pivot 5 podcast on Spotify, Apple Podcasts, and more