June Sees 357% Year-Over-Year Increase in AI Referrals to Top Websites, Surpassing 1.13 Billion

<div>
    <h2>AI Referrals Surge, Yet Google Search Dominates Traffic</h2>
    <p id="speakable-summary" class="wp-block-paragraph">While AI referrals to websites are on the rise, they still lag behind the traffic generated by Google Search. Recent insights from market intelligence firm <a target="_blank" href="https://www.similarweb.com/" rel="noreferrer noopener nofollow">Similarweb</a> reveal that AI platforms generated over 1.13 billion referrals to the top 1,000 websites globally in June, marking a staggering 357% increase since June 2024.</p>

    <h3>Google Search: The King of Referrals</h3>
    <p class="wp-block-paragraph">Despite the growth in AI-generated referrals, Google Search continues to dominate. In June 2025 alone, it accounted for an overwhelming 191 billion referrals to these sites.</p>

    <h3>The Impact on News and Media</h3>
    <p class="wp-block-paragraph">The news and media sector is particularly affected. Online publishers are bracing for a scenario dubbed "Google Zero," predicting a future where <a target="_blank" href="https://www.theverge.com/24167865/google-zero-search-crash-housefresh-ai-overviews-traffic-data-audience" rel="noreferrer noopener nofollow">Google ceases to send traffic</a> to their websites.</p>
    <p class="wp-block-paragraph">The Wall Street Journal reports alarming trends where <a target="_blank" href="https://techcrunch.com/2025/06/10/googles-ai-overviews-are-killing-traffic-for-publishers/">AI overviews are drastically reducing traffic to news sites</a>. A recent <a target="_blank" href="https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/" rel="noreferrer noopener nofollow">Pew Research Center study</a> found that AI overviews reduced click-through rates significantly. Among 69,000 Google searches surveyed, AI summaries appeared in 18% of searches, resulting in only 8% of users clicking links. In contrast, when no AI summary was present, clicks jumped to 15%.</p>

    <h3>AI Referrals in News and Media: A Rapid Rise</h3>
    <p class="wp-block-paragraph">According to Similarweb, AI referrals to news and media websites have surged by 770% since June 2024. Variations exist, as some outlets, like The New York Times, have restricted AI access due to ongoing legal disputes with OpenAI over article usage.</p>

    <h3>Top AI Referral Sources in News Media</h3>
    <p class="wp-block-paragraph">Leading the charge in AI referrals for June 2025 were several prominent news organizations: Yahoo (2.3 million), Yahoo Japan (1.9 million), Reuters (1.8 million), The Guardian (1.7 million), India Times (1.2 million), and Business Insider (1.0 million).</p>

    <h3>Understanding the Metrics: How Similarweb Analyzes Referrals</h3>
    <p class="wp-block-paragraph">Similarweb defines AI referrals as traffic directed to a domain from AI platforms such as ChatGPT, Gemini, DeepSeek, Grok, Perplexity, Claude, and Liner. Notably, ChatGPT dominates this space, generating over 80% of all AI referrals to the top 1,000 domains.</p>

    <h3>Beyond News: Categories in Focus</h3>
    <p class="wp-block-paragraph">The analysis also encompassed various other sectors including e-commerce, science and education, tech/social media, arts and entertainment, and business.</p>

    <figure class="wp-block-image aligncenter size-large">
        <img loading="lazy" decoding="async" height="614" width="680" src="https://techcrunch.com/wp-content/uploads/2025/07/top-ai-referrals-june-2025.jpg?w=680" alt="Top AI Referrals June 2025" class="wp-image-3031349" />
        <figcaption class="wp-element-caption"><strong>Image Credits:</strong> Similarweb</figcaption>
    </figure>

    <h3>E-commerce AI Referrals Surged</h3>
    <p class="wp-block-paragraph">In the e-commerce sector, Amazon led the pack with 4.5 million AI referrals, followed by Etsy (2.0 million) and eBay (1.8 million) in June.</p>

    <h3>AI Referrals in Tech and Social Media</h3>
    <p class="wp-block-paragraph">Among top tech and social platforms, Google topped the list with an impressive 53.1 million AI referrals, followed by Reddit (11.1 million), Facebook (11.0 million), GitHub (7.4 million), Microsoft (5.1 million), Canva (5.0 million), Instagram (4.7 million), LinkedIn (4.4 million), Bing (3.1 million), and Pinterest (2.5 million).</p>

    <h3>Exclusion of OpenAI in the Analysis</h3>
    <p class="wp-block-paragraph">It's important to note that the analysis did not include OpenAI's website, as a significant portion of its referrals came directly from ChatGPT.</p>

    <h3>Top AI Referral Sites Across Various Categories</h3>
    <p class="wp-block-paragraph">The top sites by AI referrals for each category featured YouTube (31.2 million), Research Gate (3.6 million), Zillow (776.2K), Europa.eu (992.9K), Wikipedia (10.8 million), NIH.gov (5.2 million), Investing.com (1.2 million), Home Depot (1.2 million), Kayak (456.5K), and Zara (325.6K).</p>
</div>

This rewritten article uses engaging headlines marked with HTML formatting and optimizes for SEO by clearly defining the main points of interest and providing an informative and structured approach to the content.

Here are five FAQs based on the significant increase in AI referrals to top websites:

FAQ 1: What does a 357% increase in AI referrals indicate?

Answer: A 357% increase in AI referrals suggests that more people and organizations are using AI-driven tools and platforms to discover and access content. This surge signifies growing trust and reliance on AI for information retrieval and decision-making.

FAQ 2: What were the total AI referrals in June?

Answer: In June, AI referrals to top websites reached 1.13 billion. This milestone reflects the rising popularity and integration of AI technologies in everyday online activities.

FAQ 3: Which industries are most impacted by the rise in AI referrals?

Answer: Industries such as e-commerce, content creation, education, and technology are particularly impacted. They are leveraging AI to enhance user experiences, improve marketing strategies, and optimize customer service.

FAQ 4: How does this increase affect website traffic?

Answer: The increase in AI referrals leads to higher website traffic, as more users are directed to sites through AI tools. This can enhance visibility, engagement, and ultimately conversions for businesses.

FAQ 5: What are the implications for digital marketing strategies?

Answer: The significant rise in AI referrals suggests that businesses may need to adjust their digital marketing strategies. Emphasizing AI optimization, understanding user behavior driven by AI tools, and creating content tailored for AI recommendations can be crucial for leveraging this trend effectively.

Source link

Meta Appoints Shengjia Zhao as Chief Scientist of AI Superintelligence Division

Meta Names Shengjia Zhao as Chief Scientist of New AI Unit

Meta CEO Mark Zuckerberg has announced that former OpenAI researcher Shengjia Zhao will lead research efforts at the company’s newly established AI unit, Meta Superintelligence Labs (MSL). Zhao has played a pivotal role in OpenAI’s notable breakthroughs, including ChatGPT and GPT-4.

Zuckerberg Celebrates Zhao’s Leadership Role

In a recent post on Threads, Zuckerberg expressed his enthusiasm: “I’m thrilled to announce that Shengjia Zhao will be the Chief Scientist of Meta Superintelligence Labs. He has been our lead scientist since the lab’s inception and co-founded it with us. With our team now forming, it’s time to formalize his leadership.”

Zhao’s Role in Shaping MSL’s Research Agenda

Zhao will spearhead the research direction for MSL under the guidance of Alexandr Wang, the recent appointee from Scale AI, who is set to head the new division.

Building a Strong Leadership Team

Strategic Hires and Research Focus

Wang, although lacking a research background, is regarded as a unique choice to lead the AI lab. The inclusion of Zhao, a respected figure in AI research, strengthens the team’s expertise. Meta has also onboarded numerous elite researchers from OpenAI, Google DeepMind, and other renowned institutions.

Zhao’s Proven Track Record in AI Innovation

Zuckerberg emphasized Zhao’s significant contributions, which include the development of a “new scaling paradigm” referenced in his work with OpenAI’s reasoning model, o1. This area is crucial for MSL as it currently lacks a competing model.

Recent Recruitment Developments

Reports indicate Zhao joined Meta alongside three other key OpenAI researchers, contributing to a robust foundation for MSL. The recruitment of Trapit Bansal and other prominent talents underlines Meta’s commitment to bolstering its AI capabilities.

Recruitment and Investment Strategies

To ensure MSL’s success, Zuckerberg is actively recruiting top talent, reportedly offering lucrative compensation packages. The company is also investing heavily in cloud computing infrastructure to facilitate ambitious AI training initiatives.

Prometheus: Meta’s Future AI Hub

By 2026, Zhao and his team will leverage Meta’s massive 1 gigawatt cloud computing cluster, Prometheus, situated in Ohio. This facility is anticipated to empower Meta to execute extensive training runs necessary for developing competitive AI models.

Looking Ahead: Collaboration Among Meta’s AI Units

With Zhao on board, Meta now has two chief AI scientists, including Yann LeCun of the FAIR lab, which focuses on long-term AI research. The collaboration between MSL and FAIR will be pivotal in shaping Meta’s AI future.

A New Era for Meta in AI Development

Overall, Meta appears to be assembling a formidable leadership team in the AI sector, positioning itself strongly against competitors like OpenAI and Google.

Here are five FAQs featuring Shengjia Zhao as the Chief Scientist of the AI Superintelligence Unit.

FAQ 1: What is the AI Superintelligence Unit?

Q: What is the AI Superintelligence Unit and what are its main objectives?

A: The AI Superintelligence Unit, led by Chief Scientist Shengjia Zhao, focuses on developing advanced AI systems that can perform tasks beyond human capabilities. Our main objectives are to ensure the safe and ethical development of superintelligent AI, explore its potential benefits, and establish guidelines for responsible integration into society.

FAQ 2: Who is Shengjia Zhao?

Q: Can you tell us about Shengjia Zhao and his role in the AI Superintelligence Unit?

A: Shengjia Zhao is the Chief Scientist of the AI Superintelligence Unit, where he leads research initiatives aimed at advancing AI technologies. With a background in machine learning and optimization, he is dedicated to ensuring that AI developments prioritize safety, ethics, and societal impact.

FAQ 3: What are the ethical considerations in developing superintelligent AI?

Q: What ethical considerations does the unit address in developing superintelligent AI?

A: The unit, under Shengjia Zhao’s leadership, addresses several ethical considerations, including fairness, transparency, accountability, and the long-term implications of superintelligent systems. We strive to create frameworks that mitigate risks while maximizing the positive impact of AI on humanity.

FAQ 4: How does the unit ensure the safety of superintelligent AI?

Q: What measures are in place to ensure the safety of superintelligent AI systems?

A: The unit conducts rigorous testing and validation of AI systems to identify potential risks. We adopt a multilayered approach that includes continuous monitoring, simulation of various scenarios, and adherence to ethical guidelines, all overseen by Shengjia Zhao to ensure that safety is a top priority.

FAQ 5: How can the public stay informed about advancements in AI?

Q: How can the public stay updated on the unit’s advancements in AI and superintelligence?

A: The public can stay informed through our official website and social media channels, where we regularly publish updates, research findings, and insights from Chief Scientist Shengjia Zhao. We also host webinars and public discussions to engage with the community and address any questions or concerns.

Source link

Chime Investor Lauren Kolodny Invests in AI to Transform Estate Processing

Lauren Kolodny: Championing Technology for Financial Inclusion

Lauren Kolodny, partner at Acrew Capital, is a staunch advocate for technology’s role in democratizing access to financial services for everyday individuals.

Backing Chime: A Bold Investment in Financial Innovation

In 2016, when the nascent neobank Chime faced skepticism from investors about its potential to serve the working class, Kolodny became the sole VC willing to invest, committing a $9 million Series A extension just as the company was on the brink of insolvency.

Reaping the Rewards: Chime’s Phenomenal Growth

That decision proved lucrative when Chime went public last month at an impressive $14.5 billion valuation.

Continuing the Mission: Investing in Consumer-Centric Tech

Kolodny, a three-time member of the Forbes Midas List, remains dedicated to funding tech solutions that empower consumers to better manage their finances.

Pioneering AI in Estate Settlements: Kolodny’s Latest Investment

Recently, she led a significant $20 million Series A investment in Alix, an innovative startup utilizing AI to streamline the estate settlement process.

A Personal Journey Fuels Innovation

Alix’s founder, Alexandra Mysoor, found inspiration after assisting a friend with settling a family estate. She shared with TechCrunch that the endeavor consumed 900 hours over 18 months, involving tedious tasks like contacting banks to transfer assets and locating various accounts.

Transforming an Archaic Process

“I was shocked at how complicated this process was,” said Mysoor. “It’s paper-driven and outdated. You’re searching for to-do lists that aren’t helpful and contacting attorneys who provide only a fraction of the needed work, charging exorbitant fees.”

Join us at the TechCrunch event!

San Francisco
|
October 27-29, 2025

Revolutionizing Estate Settlement with AI

This experience led Mysoor to realize that many labor-intensive elements of trust administration, such as document management and communication with banks, can be efficiently managed by AI technologies.

Kolodny’s Vision: Solving a Pressing Problem

Upon meeting Mysoor and understanding Alix’s mission, Kolodny recognized a significant issue that she couldn’t shake from her mind. As economists predict the transfer of trillions of dollars to younger generations, the burdensome paperwork for estate settlements largely falls on grieving families.

A Gap in the Market for Estate Services

Kolodny found that while some startups, such as Empathy, assist with account closures during bereavement, no companies provided end-to-end estate settlement solutions.

Aha Moment: The Need for a Comprehensive Solution

“How is it that such a complicated process, which requires extensive project management, lacks meaningful solutions?” Kolodny remarked. “It was a true ‘aha’ moment for me. This is exactly the type of challenge AI can tackle.”

Alix: Democratizing Financial Services

Kolodny believes that Alix could be among the first wave of AI-powered startups that will democratize financial and administrative services that were historically reserved for the wealthy.

Transparent Pricing: Alix’s Approach to Fees

Alix charges a fee of 1% of an estate’s total value. For inheritances below $1 million, clients can anticipate costs ranging between $9,000 and $12,000, depending on the estate’s complexity.

Here are five FAQs based on the topic of Lauren Kolodny and her investment in AI to revolutionize estate processing:

FAQ 1: Who is Lauren Kolodny?

Answer: Lauren Kolodny is a prominent backer in the tech industry, known for her investment in innovative startups. She has a strong focus on leveraging artificial intelligence to transform traditional processes, including estate processing.

FAQ 2: What is the significance of using AI in estate processing?

Answer: AI can streamline and automate various aspects of estate processing, improving efficiency and accuracy. This includes tasks such as document management, data analysis, and decision-making, ultimately making the process faster and reducing costs.

FAQ 3: How does Lauren Kolodny believe AI will change the estate industry?

Answer: Kolodny believes that AI will revolutionize the estate industry by providing tools that enhance transparency, improve speed, and reduce human error. This technological shift can also lead to better user experiences for clients navigating estate services.

FAQ 4: What challenges might arise with the integration of AI in estate processing?

Answer: Challenges may include data privacy concerns, the need for regulatory compliance, and the potential for resistance from traditional stakeholders. Additionally, there may be technical hurdles in implementing AI systems effectively within existing frameworks.

FAQ 5: How can startups in the estate processing field benefit from AI?

Answer: Startups can leverage AI to differentiate themselves in a competitive market by offering innovative solutions that simplify processes, enhance customer interactions, and provide valuable insights through data analytics. This can lead to increased customer satisfaction and potentially higher revenue.

Source link

AI Coding Challenge Unveils Initial Results – and They’re Not Encouraging

A New AI Coding Challenge Crowned Its First Winner, Setting New Standards for AI Software Engineering

A groundbreaking AI coding competition has unveiled its inaugural champion, raising the benchmark for AI-driven software engineers.

Eduardo Rocha de Andrade Claims the K Prize

On Wednesday at 5 PM PST, the Laude Institute, a nonprofit organization, announced the first winner of the K Prize—a multi-round AI coding challenge initiated by Databricks and Perplexity co-founder Andy Konwinski. The victor, Eduardo Rocha de Andrade, a Brazilian prompt engineer, will take home a prize of $50,000. Surprisingly, he secured the win by answering only 7.5% of the test questions correctly.

A Challenging Benchmark for AI Models

“We’re pleased to have established a benchmark that is genuinely challenging,” Konwinski stated. He emphasized that benchmarks should demand high standards if they are to be meaningful. He further noted, “Scores might differ if the larger labs participated with their top models. But that’s precisely the intention. The K Prize operates offline with limited computational resources, giving preference to smaller, open models. I find that exciting—it levels the playing field.”

Future Incentives for Open-Source Models

Konwinski has committed $1 million to the first open-source model that achieves a score above 90% on the K Prize assessment.

The K Prize’s Unique Approach

Similar to the renowned SWE-Bench system, the K Prize evaluates models based on GitHub issues as a way to assess their ability to tackle real-world programming challenges. However, the K Prize sets itself apart by employing a “contamination-free version of SWE-Bench,” utilizing a timed entry system to prevent any benchmark-specific training. For the initial round, models were due by March 12th, and the organizers constructed the test using only GitHub issues flagged after that date.

A Stark Contrast in Scoring

The 7.5% winning score contrasts sharply with SWE-Bench, which reports a top score of 75% on its easier ‘Verified’ test and 34% on its more challenging ‘Full’ test. While Konwinski remains uncertain if this discrepancy is due to contamination in SWE-Bench or the complexity of gathering new GitHub issues, he anticipates the K Prize will provide clarity soon.

Future Developments and Evolving Standards

“As we conduct more rounds, we’ll gain better insight,” he told TechCrunch, “as we expect competitors to adapt to the evolving landscape every few months.”

Join us at the upcoming TechCrunch event

San Francisco
|
October 27-29, 2025

Addressing AI’s Evaluation Challenges

While it may seem unexpected for AI coding tools to struggle, critics argue that initiatives like the K Prize are vital for addressing AI’s escalating evaluation dilemma.

Advancing Benchmarking Methodologies

“I’m optimistic about developing new tests for existing benchmarks,” says Princeton researcher Sayash Kapoor, who proposed a similar concept in a recent paper. “Without these experiments, we can’t definitively ascertain if the problem lies in contamination or merely targeting the SWE-Bench leaderboard with human input.”

A Reality Check for AI Aspirations

For Konwinski, this challenge is not just about creating a better benchmark—it’s a call to action for the entire industry. “If you listen to the hype, you’d think AI doctors, lawyers, and software engineers should already be here, but that’s simply not the reality,” he asserts. “If we can’t surpass 10% on a contamination-free SWE-Bench, that serves as a stark reality check for me.”

Here are five FAQs about the recent AI coding challenge results:

FAQ 1: What was the AI coding challenge about?

Answer: The AI coding challenge aimed to evaluate the performance and capabilities of advanced AI models in solving complex coding tasks. Participants submitted their solutions, which were then assessed for accuracy, efficiency, and creativity.

FAQ 2: What were the results of the challenge?

Answer: The first results indicated that the AI models struggled significantly with coding tasks. Many submissions lacked the expected quality and often failed to meet the basic requirements of the challenges, highlighting limitations in current AI capabilities.

FAQ 3: What factors contributed to the poor results?

Answer: Several factors contributed to the disappointing outcomes, including ambiguity in problem statements, limitations in the training data, and challenges in understanding nuanced coding concepts. Additionally, the complexity of the tasks might have exceeded the current capabilities of the AI models.

FAQ 4: How will the organizers address the issues highlighted by the results?

Answer: The organizers plan to analyze the submissions in more detail, gathering feedback from participants and experts to improve future challenges. They aim to revise problem statements for clarity and consider introducing more comprehensive training resources for participants.

FAQ 5: What is the outlook for future AI coding challenges?

Answer: While the initial results were discouraging, the outlook remains positive. The organizers believe that with iterative improvements and increased collaboration within the AI community, future challenges can lead to better performance and advancements in AI coding capabilities.

Source link

Amazon Purchases Bee: The AI Wearable That Captures Every Conversation

<div>
    <h2>Amazon Acquires AI Wearables Startup Bee: A New Frontier in Wearable Technology</h2>

    <p id="speakable-summary" class="wp-block-paragraph">In a recent announcement, Amazon has confirmed its acquisition of the AI wearables startup Bee, as revealed by Bee co-founder Maria de Lourdes Zollo on LinkedIn. Although the deal has yet to formally close, it signifies a promising step for Amazon into the realm of AI-driven wearable devices.</p>

    <h3>Bee’s Innovative Approach to Wearable Technology</h3>
    <p class="wp-block-paragraph">Last year, Bee raised $7 million to develop its innovative products, which include a Fitbit-like bracelet priced at $49.99, alongside a subscription option of $19 per month, and an Apple Watch app. The device is designed to record everything it hears (unless muted), transforming conversations into actionable reminders and to-do lists for users.</p>

    <h3>A Vision for Seamless Connectivity</h3>
    <p class="wp-block-paragraph">Zollo shared with TechCrunch that Bee’s ambition is to create a "cloud phone," enabling their personal device to access user accounts and notifications for timely reminders and messaging capabilities.</p>

    <h3>Transforming AI Interaction</h3>
    <p class="wp-block-paragraph">Bee aims to provide a personal, ambient intelligence that feels more like a trusted companion, helping users reflect, remember, and navigate daily life with ease.</p>

    <h3>Market Competitors and Pricing Strategy</h3>
    <p class="wp-block-paragraph">While other companies like Rabbit and Humane AI have ventured into similar AI-enabled wearables, they have struggled to achieve significant success. Bee’s affordability at $50 makes it an attractive option for consumers wary of higher financial commitments, especially compared to the now-defunct Humane AI Pin, which cost $499.</p>

    <h3>Integration into Amazon: Opportunities for Bee Employees</h3>
    <p class="wp-block-paragraph">An Amazon spokesperson confirmed that employees from Bee have received offers to join the tech giant, paving the way for further innovation within its ecosystem.</p>

    <h3>Amazon's Strategic Move into Wearable AI</h3>
    <p class="wp-block-paragraph">This acquisition highlights Amazon’s strategic interest in wearable AI technology, branching out from its well-known voice-activated devices like Echo speakers. In parallel, other tech entities like OpenAI and Meta are also making strides in this space.</p>

    <h3>Security and Privacy Concerns in AI Wearables</h3>
    <p class="wp-block-paragraph">Despite the potential benefits, security and privacy risks loom over AI wearables that capture ambient audio. Varying policies across companies dictate how recordings are handled, raising concerns for users.</p>

    <h3>Bee's Commitment to User Privacy</h3>
    <p class="wp-block-paragraph">Bee's current privacy policies allow users to delete their data at will, claiming that audio recordings are not stored or used for AI training. The app does retain certain learned data to enhance its assistant functionality, ensuring a personalized experience.</p>

    <h3>Looking Ahead: Potential Changes Under Amazon's Ownership</h3>
    <p class="wp-block-paragraph">It remains to be seen how Bee’s privacy policies will evolve once integrated into Amazon, especially considering Amazon’s mixed track record regarding user data management.</p>

    <h3>Historical Privacy Issues at Amazon</h3>
    <p class="wp-block-paragraph">Past incidents reveal serious privacy concerns, including Amazon's sharing of Ring camera footage with law enforcement without user consent and agreements addressing lax security practices within its devices.</p>
</div>

This revised article features improved structure and SEO optimization, including engaging headlines and a clear hierarchy, making it both informative and search-friendly.

Certainly! Here are five FAQs regarding Amazon’s acquisition of Bee, the AI wearable device:

FAQ 1: What is Bee and how does it work?

Answer: Bee is an AI-powered wearable device that records everything you say. It uses advanced voice recognition technology to capture audio in real time, allowing users to easily access and review conversations or notes later. The device can connect seamlessly with other devices, making it ideal for personal use or professional settings.

FAQ 2: Why did Amazon acquire Bee?

Answer: Amazon acquired Bee to enhance its portfolio in the wearable technology sector and to improve its voice recognition and AI capabilities. By integrating Bee’s technology, Amazon aims to deepen its focus on AI-driven products, potentially expanding features in existing devices like Alexa.

FAQ 3: What privacy measures are in place for users of Bee?

Answer: Bee prioritizes user privacy by employing robust encryption and data security protocols. Users have control over their recordings, with options to delete, export, or manage recorded data. Amazon also adheres to strict compliance with privacy regulations, ensuring that user data is handled responsibly.

FAQ 4: How will this acquisition affect existing Bee users?

Answer: Existing Bee users can expect continued support and updates for their devices. Additionally, with Amazon’s backing, there may be new features, integrations, or enhancements introduced, making the user experience even better.

FAQ 5: Will Amazon integrate Bee’s technology into other products?

Answer: Yes, it is likely that Amazon will integrate Bee’s AI technology into its existing products, particularly within its Echo and Alexa user ecosystem. This could lead to enhanced voice interaction capabilities and new functionalities, boosting the overall user experience.

Feel free to ask if you have more specific questions or need further information!

Source link

Latent Labs Introduces Web-Based AI Model to Make Protein Design Accessible to All

Latent Labs Unveils Groundbreaking AI Model for Programmable Biology

Six months after emerging from stealth mode with $50 million in funding, Latent Labs has launched a revolutionary web-based AI model aimed at programming biology.

Achieving State-of-the-Art Proteins with AI

According to Simon Kohl, CEO and founder of Latent Labs and former co-lead of DeepMind’s AlphaFold protein design team, the Latent Labs model has “achieved state-of-the-art on different metrics” during tests of the proteins created within a physical lab. The term “state-of-the-art,” or SOTA, is often used to denote the highest level of performance in AI for a given task.

Innovative Assessment Methods

“We have computational ways of assessing how good the designs are,” Kohl told TechCrunch, highlighting that a significant percentage of proteins generated by the model are expected to be viable in laboratory tests.

Introducing LatentX: A New Frontier in Protein Design

LatentX, the company’s foundational biology model, allows academic institutions, biotech startups, and pharmaceutical companies to design novel proteins directly from their browser using natural language.

Pushing Beyond Nature’s Limitations

Unlike existing biological frameworks, LatentX can create entirely new molecular designs, including nanobodies and antibodies with exact atomic configurations, significantly accelerating the development of new therapeutics.

Distinct from AlphaFold

Kohl emphasizes that LatentX’s ability to design new proteins sets it apart from AlphaFold: “AlphaFold is a model for protein structure prediction, enabling visualization of existing structures, but it does not facilitate the generation of new proteins.”

Licensing Model to Democratize AI Access

In contrast to other AI-driven drug discovery companies such as Xaira, Recursion, and DeepMind spinout Isomorphic Labs, Latent Labs adopts a licensing approach that allows external organizations to utilize its model.

Future Monetization Plans

While LatentX is currently available for free, Kohl indicated that the company plans to charge for advanced features and capabilities as they are rolled out in the future.

Open-Source Collaboration in Drug Discovery

Other firms providing open-source AI foundational models for drug discovery include Chai Discovery and EvolutionaryScale.

Backed by Industry Leaders

Latent Labs benefits from the backing of notable investors, including Radical Ventures, Sofinnova Partners, Google Chief Scientist Jeff Dean, Anthropic CEO Dario Amodei, and Eleven Labs CEO Mati Staniszewski.

Here are five FAQs with answers regarding the launch of Latent Labs’ web-based AI model aimed at democratizing protein design:

1. What is the purpose of Latent Labs’ new AI model?

Latent Labs’ new web-based AI model aims to democratize protein design, making advanced biotechnological tools accessible to researchers, companies, and enthusiasts. This model simplifies the process of designing proteins, which can have applications in medicine, environmental science, and biotechnology.

2. How does the AI model work?

The AI model utilizes machine learning algorithms trained on extensive protein data to predict and generate novel protein structures and functions. Users can input specific parameters, and the model will provide optimized designs that meet various criteria, streamlining the experimental process.

3. Who can use this web-based AI model?

The platform is designed for a wide range of users, including academic researchers, biotech companies, students, and hobbyists interested in protein engineering. Its accessibility aims to empower individuals and organizations without extensive resources or expertise in computational biology.

4. What are the potential applications of the designed proteins?

The proteins designed using this AI model can serve various purposes, including therapeutic applications (such as drug development), industrial uses (like enzyme production for sustainable processes), and research purposes (to study protein functions and interactions).

5. Is there any cost associated with using the AI model?

While specific pricing details may vary, Latent Labs intends to offer free or affordable access options to ensure that the technology is widely available. Users should check the Latent Labs website for the latest information on access, subscription plans, and any associated costs.

Source link

Why Cartken Shifted Its Focus from Last-Mile Delivery to Industrial Robotics

Cartken Shifts Focus: From Food Delivery to Industrial Robotics

Autonomous robotics startup Cartken, recognized for its innovative food delivery robots operating on college campuses and in Tokyo, is now venturing into the industrial sector.

Identifying Industrial Opportunities

Christian Bersch, co-founder and CEO of Cartken, shared with TechCrunch that the idea to adapt delivery robots for industrial applications was a consideration from the startup’s inception. As companies began expressing interest, Cartken explored this new avenue further.

Revealing Market Demand

“We discovered a significant need for onsite industrial solutions,” Bersch, a former Google engineer behind the Bookbot project, explained. “Companies could benefit from optimizing material and production flows.”

Initial Success with ZF Lifetec

In 2023, Cartken secured its first major industrial client, ZF Lifetec, a German manufacturer. Initially, ZF utilized Cartken’s existing delivery robot, the Courier, which can carry up to 44 pounds and resembles a wheeled cooler.

Transitioning Robots to Industrial Use

“Our food delivery robot started transporting production samples, quickly becoming our busiest unit,” Bersch said. “This success made us realize the viable market need, prompting us to target this sector more aggressively.”

Ongoing Food Delivery Partnerships

While Cartken continues to expand its sidewalk delivery business—with partnerships with Uber Eats and GrubHub for last-mile operations in U.S. colleges and Japan—its industrial success encouraged a broader business model exploration.

Seamless Adaptation to Industrial Settings

Bersch noted that adapting the robots for industrial roles was straightforward. The AI, trained on years of food delivery data, allows the robots to navigate diverse terrains and weather conditions.

Cartken Hauler in a warehouse
Image Credits:Cartken

Expanding the Robot Fleet

Having raised over $20 million from investors such as 468 Capital and Incubate Fund, Cartken is expanding its robotic fleet to cater to industrial needs. Earlier this year, it launched the Cartken Hauler, capable of carrying up to 660 pounds, as well as the Cartken Runner for indoor deliveries and plans for a robotic forklift.

Advanced Navigation and Flexibility

Bersch stated, “Our navigation system is customizable for different robot sizes, and the AI from food delivery is directly applicable to these new models.”

Deepening Partnerships with Mitsubishi

Recently, Cartken reinforced its partnership with Mitsubishi, which had initially helped in obtaining necessary certifications for its delivery robots in Tokyo. Melco Mobility Solutions, a Mitsubishi subsidiary, is set to acquire nearly 100 Cartken Hauler robots for Japanese industrial applications.

Broader Applications in Various Industries

“We see growing interest from diverse industrial sectors, including automotive, pharmaceuticals, and chemicals,” Bersch noted. “These companies often need to transport items between buildings, traditionally done by hand or with small forklifts—this is where we come in.”

Future of Food Delivery

While Cartken will maintain its food and consumer last-mile delivery services, Bersch indicated that expansion in this area isn’t in the immediate plans, emphasizing ongoing testing of new features within existing last-mile delivery routes.

Sure! Here are five FAQs regarding Cartken’s shift from last-mile delivery to industrial robots:

FAQ 1: Why did Cartken pivot from last-mile delivery to industrial robots?

Answer: Cartken identified that while last-mile delivery was a growing sector, the increasing demand for industrial automation offered a broader market opportunity. By focusing on industrial robots, they can leverage their technology to serve various industries, leading to increased efficiency and new applications.

FAQ 2: What advantages does Cartken see in industrial robots compared to last-mile delivery?

Answer: Industrial robots can enhance operational efficiency, reduce labor costs, and improve safety in manufacturing and warehousing environments. The scalability and adaptability of these robots allow Cartken to address a wider range of industrial challenges, making this pivot strategically advantageous.

FAQ 3: How does this pivot align with broader industry trends?

Answer: The global push for automation and digital transformation in industries is rapidly increasing. By moving into industrial robotics, Cartken aligns itself with this trend, ensuring it remains competitive and relevant, while catering to businesses seeking to automate and optimize their operations.

FAQ 4: Will Cartken still engage in last-mile delivery or focus solely on industrial robots?

Answer: While Cartken’s primary focus has shifted to industrial robotics, they may still explore opportunities in last-mile delivery if synergies arise. The pivot, however, indicates a strong commitment to capitalizing on the growth and potential of the industrial sector.

FAQ 5: What types of industries can benefit from Cartken’s industrial robots?

Answer: Cartken’s industrial robots are designed for a variety of sectors, including manufacturing, logistics, and warehousing. They can be utilized for tasks such as material handling, assembly, and inventory management, offering diverse applications across different industrial settings.

Source link

Windsurf CEO Discusses the ‘Very Bleak’ Atmosphere Prior to the Cognition Deal

Windsurf Acquired by Cognition: A Tale of Transition and Turmoil

Following the acquisition of AI coding startup Windsurf by Cognition, executive Jeff Wang took to X to shed light on the challenges surrounding the deal.

Failed Talks with OpenAI Opened New Doors

Windsurf was initially in acquisition talks with OpenAI, but that deal collapsed. Instead, Google DeepMind hired CEO Varun Mohan and other key personnel from Windsurf. Reports indicate Google will license Windsurf’s technology for $2.4 billion but will not acquire the company outright.

The Rise of “Reverse Acquihires”

This incident highlights a growing trend of “reverse acquihires,” where major tech firms hire key members from startups to mitigate antitrust concerns while licensing their technologies rather than executing full acquisitions.

Impact on Employees Left Behind

This raises a critical question: What happens to the startups and their employees once top talent departs? In a recent episode of Equity, a founder likened leaving executives to a captain abandoning ship in turbulent waters.

Windsurf’s Leadership Step Up Amidst Uncertainty

After Mohan’s exit, Wang, previously the head of business, took over as interim CEO. He expressed sympathy for Mohan and Chen, recognizing the difficulty of their situation.

All-Hands Meeting Reveals Employee Sentiments

During a company-wide meeting on June 11, expectations were high for news about the OpenAI deal. Instead, Wang had to share the disappointing Google acquisition and the departure of key figures. “The mood was very bleak,” he reflected. “Some were upset about financial outcomes, while others were anxious about the future; a few were in tears.”

Potential for Recovery

Despite setbacks, Wang believes Windsurf still has significant assets, including intellectual property and talented personnel, to pursue further investment, a sale, or continuing operations.

Negotiations with Cognition Begin

That same evening, Wang was in discussions with Cognition’s Scott Wu and Russell Kaplan. Following a frantic weekend of negotiations, they kept interest from other potential suitors in mind while also addressing the needs of Windsurf’s remaining engineers.

A Strategic Fit for Future Growth

Wang argued that Cognition and Windsurf make a great partnership due to complementary strengths. “Cognition had overinvested in engineering but underinvested in go-to-market and marketing,” he explained, adding that Windsurf possesses world-class talent in these areas.

Commitments to Employee Welfare

Wang noted a focus on ensuring the welfare of Windsurf’s employees was paramount during negotiations, resulting in a deal structure that includes payouts for all staff, the waiving of cliffs, and accelerated vesting for Windsurf equity.

A Rollercoaster Weekend: From Fear to Hope

The acquisition agreement was finalized at 9:30 AM on Monday, announced to the team shortly after, and disclosed to the public not long thereafter. In an interview with Bloomberg, Wang described the tumultuous Friday as “probably the worst day of 250 people’s lives,” followed by what felt like “probably the best day.”

Here are five FAQs with answers based on the scenario involving a Windsurf CEO discussing the mood before the Cognition deal:

FAQ 1: What prompted the CEO to describe the mood as "very bleak" before the Cognition deal?

Answer: The CEO felt the mood was "very bleak" due to a combination of challenging market conditions, declining sales, and a lack of innovative product development, which put pressure on the company’s performance and future growth.

FAQ 2: What was the significance of the Cognition deal for Windsurf?

Answer: The Cognition deal was significant because it represented a strategic partnership that could revitalize Windsurf’s product line, drive innovation, and improve market positioning, ultimately paving the way for recovery and growth.

FAQ 3: How did the CEO feel about the future after the Cognition deal was finalized?

Answer: After finalizing the Cognition deal, the CEO expressed optimism about the future. They believed the partnership would bring new resources, innovative ideas, and a renewed sense of direction for the company.

FAQ 4: What steps is Windsurf taking post-deal to improve its market outlook?

Answer: Windsurf is focusing on integrating Cognition’s capabilities, investing in research and development, and enhancing marketing strategies to better engage consumers and expand its market presence.

FAQ 5: How does the CEO plan to address the "bleak" mood among employees following the deal?

Answer: To address the mood among employees, the CEO plans to enhance internal communication, provide updates on progress and improvements, and foster a culture of openness and collaboration to rebuild morale and encourage a collective focus on future goals.

Source link

Benchmark Negotiating Series A Investment for Greptile, Valuing AI Code Reviewer at $180M, Sources Indicate

Greptile: The AI-Powered Code Review Startup Eyeing $30M Series A

Greptile, an innovative startup leveraging AI for code reviews, is in the process of securing a $30 million Series A funding round at a valuation of $180 million, led by Benchmark partner Eric Vishria. However, sources indicate that the deal is not yet finalized, and terms may be subject to change.

Founding and Early Success

Founded by Daksh Gupta shortly after graduating from Georgia Tech in 2023, Greptile gained momentum through its participation in Y Combinator’s winter 2024 cohort. Following this, they successfully raised a $4 million seed round led by Initialized Capital.

AI Code Review Technology

Gupta explained to TechCrunch that Greptile’s AI bot functions like an experienced colleague, adept at understanding the intricacies of a customer’s code. This capability enables it to identify bugs and issues that might elude human reviewers.

Operating in a Competitive Landscape

The space for AI code review solutions is highly competitive. Notable rivals include Graphite, which raised $52 million Series B earlier this year led by Accel, and Coderabbit, which secured a $16 million Series A from CRV last year.

Work Culture and Employee Demands

The fierce competition has resulted in Greptile implementing demanding work hours for its staff. Gupta controversially shared on X that Greptile “offers no work-life balance,” with employees typically clocking in from 9 AM to 11 PM, including weekends.

Maximizing Effort in a Cutthroat Environment

After his post gained attention, Gupta remarked to various media outlets that excelling in such a competitive field requires unmatched dedication from every team member. “No one cares about the third-best company,” he stated in an interview with Inc., stressing the importance of total commitment over partial effort.

Looking Ahead: The Impact of Series A Funding

Despite its challenging work culture, attracting a prestigious venture capital firm like Benchmark at a robust valuation could significantly bolster Greptile’s future.

Both Greptile and Benchmark have not responded to requests for comment.

Techcrunch Event

San Francisco
|
October 27-29, 2025

Here are five FAQs based on the provided benchmark regarding Greptile’s Series A funding and AI-code reviewer valuation:

FAQ 1: What is Greptile, and what does it offer?

Answer: Greptile is an innovative technology company that specializes in AI-driven code review solutions. By leveraging artificial intelligence, Greptile enhances the code development process, ensuring higher quality and efficiency in software projects.

FAQ 2: What is the significance of the $180 million valuation?

Answer: The $180 million valuation underscores Greptile’s potential impact in the software development industry. It reflects investor confidence in the company’s technology, market position, and growth prospects, especially within the rapidly evolving AI sector.

FAQ 3: What are the expected outcomes of the Series A funding?

Answer: The Series A funding is expected to accelerate product development, enhance marketing efforts, and expand Greptile’s team. This growth phase aims to solidify its market presence and improve user adoption of its AI code review tools.

FAQ 4: Why is AI-driven code review important for developers?

Answer: AI-driven code review automates and improves the code review process by detecting bugs and suggesting improvements at a faster pace than traditional methods. This leads to higher code quality, reduced development time, and allows developers to focus on more complex tasks.

FAQ 5: What investors are involved in this Series A funding round?

Answer: While specific investors may not be publicly disclosed, Series A funding usually involves venture capital firms that specialize in technology and AI investments. These investors are attracted to companies with strong growth potential and innovative solutions, like Greptile’s AI code reviewer.

Source link

Perplexity Views India as a Strategic Advantage in the Competition with OpenAI

Perplexity’s Strategic Expansion into India’s AI Market

While OpenAI dominates the U.S. AI landscape, Perplexity is quietly embarking on an ambitious journey in India. The search-centric AI startup is successfully attracting millions of users in the world’s second-largest internet and smartphone market, gearing up for significant mass-market presence.

Exclusive Partnership with Bharti Airtel

In a significant move this week, Perplexity collaborated with Bharti Airtel, India’s second-largest telecom operator, to provide a complimentary 12-month Perplexity Pro subscription—valued at $200—to all 360 million Airtel subscribers. Airtel confirmed to TechCrunch that this exclusive deal prohibits other telecom providers from offering Perplexity’s services to their customers.

Global Expansion Strategy: Building Volume

The Airtel collaboration marks one of Perplexity’s most noteworthy actions as part of a broader global expansion strategy, which includes alliances with over 25 telecom companies worldwide, such as those recently established with SoftBank in Japan and SK Telecom in South Korea. Given India’s massive population, the startup sees unparalleled opportunities for growth not found in other regions.

Impressive Growth Metrics in India

Perplexity is witnessing remarkable success in India, with downloads skyrocketing 600% year-over-year to 2.8 million in Q2, according to exclusive Sensor Tower data shared with TechCrunch. In comparison, OpenAI’s ChatGPT observed a 587% rise, totaling 46.7 million downloads during the same timeframe.

The uptick in active users mirrors this growth; Perplexity’s monthly active users (MAUs) surged 640% year-over-year in Q2, while ChatGPT’s grew by 350%. India ranked as Perplexity’s largest MAU market last quarter, per Sensor Tower, although ChatGPT still holds a substantial lead with 19.8 million MAUs compared to Perplexity’s 3.7 million.

Comparison of Perplexity and ChatGPT growth in India
Image Credits:Jagmeet Singh / TechCrunch

Leveraging India’s Unique Market Dynamics

Building on prior partnerships, Perplexity aims to utilize India’s substantial user base to leapfrog over mature Western markets where OpenAI currently dominates in paid subscriptions. Earlier this year, it also partnered with Paytm, a leading Indian fintech, to integrate its AI-powered search into the Paytm app, boasting over 500 million downloads and ranking among the top three apps within India’s Unified Payment Interface.

CEO Aravind Srinivas’s Commitment to India

Perplexity’s CEO, Aravind Srinivas, has taken proactive steps to bolster the company’s presence in India. In January, he announced plans to hire an Indian executive, later pausing the initiative due to an “overwhelming” response to the job listing. He has since committed to a $1 million investment and dedicating five hours per week to boost AI initiatives in India.

Targeting Students and Tech-Savvy Users

Sources indicate Perplexity is also considering providing its AI search engine to Indian students to broaden its reach further.

Capitalizing on Limited Local Competition

A key factor for Perplexity’s focus on India is the relatively limited number of local AI startups in the AI search sector, paired with a large base of tech-savvy users. This dynamic has prompted even Google to introduce AI-focused search tools like AI Mode in India ahead of many other regions.

Monetization Challenges and Opportunities

Despite its growth, Perplexity faces significant challenges in monetizing its user base. It still trails far behind ChatGPT globally in revenue generation, despite both platforms offering a comparable $20 monthly subscription. ChatGPT’s in-app purchase revenue reached $773 million in Q2, reflecting a 731% year-over-year increase, while Perplexity reported an increase to $8 million, a 300% growth, as per Sensor Tower data.

Revenue comparison of Perplexity and ChatGPT
Image Credits:Jagmeet Singh / TechCrunch

Future Potential in an Evolving Market

In India, where consumers are known to be price-sensitive, Perplexity’s monetization strategy will require careful consideration. However, promising indicators exist. ChatGPT’s in-app purchase revenue saw an 800% year-over-year increase to $9 million in India during Q2. While Perplexity hasn’t captured notable in-app revenue yet, partnerships like the one with Airtel present opportunities for subscription growth in the short term.

Attracting Investment through User Growth

Strategic alliances in markets like India could position Perplexity favorably in front of investors who prioritize user growth and geographic diversification. To ensure sustained support, the startup must demonstrate the ability to convert its expanding user base into revenue streams.

Srinivas did not respond to requests for comment.

FAQs about Perplexity and Its Strategy in India

  1. What is Perplexity’s main goal in India?

    • Perplexity aims to establish a strong foothold in India’s rapidly growing technology market. By leveraging local talent and resources, the company seeks to innovate and compete effectively against giants like OpenAI.

  2. How does Perplexity plan to compete with OpenAI?

    • Perplexity is focusing on developing unique AI solutions tailored to the needs of the Indian market. This includes enhancing natural language processing capabilities and integrating AI into everyday applications, providing a niche that differentiates it from OpenAI’s offerings.

  3. What advantages does India offer to Perplexity?

    • India provides a vast pool of skilled tech professionals, a burgeoning startup ecosystem, and increasing investment in AI and technology sectors. This environment fosters innovation and collaboration, allowing Perplexity to accelerate its growth.

  4. Are there specific sectors in India where Perplexity will focus its efforts?

    • Yes, Perplexity plans to concentrate on sectors like healthcare, education, and finance, where AI can drive significant improvements and create impactful solutions that align with local needs.
  5. What are the potential challenges Perplexity might face in India?
    • Challenges include intense competition from both local startups and established global players, navigating regulatory landscapes, and adapting to diverse consumer preferences across regions in India.

Source link