Artificial Intelligence News Live: AI code reviewer Greptile reportedly in talks to raise $30M at a $180M valuation

AI code review startup Greptile is in discussions for a $30 million Series A funding round that would be led by venture capital firm Benchmark, according to a report in TechCrunch. The potential investment would value the Y Combinator alumnus company at $180 million post-money. The report, which cites unnamed sources, indicates the deal is in advanced stages but not yet finalized, reflecting strong investor interest in specialized AI tools for software developers.

Y Combinator-backed startup Pig.dev has pivoted away from its initial goal of developing autonomous AI agents capable of controlling the Windows desktop environment, TechCrunch says in a report. The company's decision highlights the significant technical difficulties and practical barriers involved in creating reliable AI that can interact with complex operating systems. These challenges include issues with perception, planning, and error handling, which persist even with the use of powerful large language models, underscoring the complexity of building truly general-purpose agents.

Algorithmic rent-setting software company RealPage has announced its acquisition of Livble, a service that allows tenants to pay rent in installments for a fee of $30 to $40 per loan, according to a report in The Verge. The acquisition comes as RealPage faces a major antitrust lawsuit from the US Department of Justice, which alleges the company used its algorithm and nonpublic competitor data to monopolize the market for rental pricing software. George Slover, senior counsel at the Center for Democracy and Technology, noted the move could draw further review from the Justice Department. The company's use of algorithms has also previously drawn scrutiny from US senators who questioned its potential connection to a scrapped proposal for a moratorium on AI regulation.

Anysphere, the company behind the AI-first code editor Cursor, has acquired enterprise startup Koala in a strategic move to bolster its competition against Microsoft's GitHub Copilot, according to a report in TechCrunch. The acquisition is aimed at securing top talent from the AI enterprise sector, positioning Anysphere to enhance its offerings and compete more directly in the growing market for AI-powered developer tools.

Search engine DuckDuckGo has launched a new feature that allows users to hide AI-generated images from their search results, according to a report in TechCrunch. This development comes amid a significant increase in synthetic media online and growing user demand for more control over content quality and authenticity. The new tool specifically targets visual content, enabling users to filter out images identified as being created by artificial intelligence, thereby addressing concerns about visual clutter and the proliferation of low-quality AI content.

Higher education institutions are grappling with the rapid pace of change brought on by generative AI, forcing them to revise academic integrity policies and adapt teaching methods, according to a report in the Financial Times. As universities work to formulate guidelines and train faculty, technology companies are quickly moving to develop and offer custom AI-powered learning products. This dynamic is reshaping the educational landscape, presenting students with new tools for personalized learning but also new challenges related to the ethical use of AI and academic standards.

Major investment firms, including BlackRock and Blackstone, are proposing to acquire energy companies to capitalize on the immense and growing power demands of the artificial intelligence industry, according to a report in the New York Times. The proposed acquisitions target regulated utilities in states such as Minnesota, New Mexico, and Texas, where the need for reliable electricity for AI data centers is surging. Consumer advocacy groups have voiced opposition, arguing that such takeovers by private equity could lead to higher electricity rates and prioritize the energy needs of large data centers over the affordability and reliability of service for residential customers.

Indie filmmakers known as the Dor Brothers are gaining widespread attention for creating viral videos produced entirely with artificial intelligence, The New York Times says in a report. Their work exemplifies a growing trend where generative AI tools are making high-quality video production accessible to smaller creators, lowering costs and technical barriers traditionally associated with complex visual content. This approach allows independent studios to compete with larger productions, though the process still relies heavily on human creativity for prompting, curating, and editing the final output.

A coalition of funders, including the Ballmer Group, will invest $1 billion over 15 years to develop artificial intelligence tools for frontline workers such as public defenders, parole officers, and social workers, AP said in a report. The funders announced the creation of a new entity, NextLadder Ventures, which will provide grants and investments to both non-profit and for-profit organizations. The initiative aims to develop tools that help public service professionals manage large caseloads and limited resources to better assist Americans in precarious situations.

Gaming platform Roblox is introducing a new age verification system that uses artificial intelligence to perform age estimation by scanning video selfies from teenage users, according to a report in Wired. The system is a component of a new feature called 'Trusted Connections,' which will allow verified users to access less filtered chat when communicating with people they know. The AI's function is to assess an age range and does not perform facial recognition to uniquely identify the user.

Artificial intelligence companies are “fundamentally unprepared” for the societal consequences of developing systems with human-level intellectual capabilities, according to a report in The Guardian. The Future of Life Institute (FLI), an AI safety organization, found that firms pursuing artificial general intelligence lack credible plans for safety. In its AI safety index, the institute noted that no company scored higher than a 'D' for “existential safety planning,” highlighting a significant gap in readiness for potential risks.

New research indicates that one of the largest open-source AI training sets, DataComp CommonPool, likely includes millions of examples of sensitive personal data, according to a report in MIT Technology Review. Investigators found thousands of images containing personally identifiable information—including passports, credit cards, birth certificates, and identifiable faces—within just a small subset of the dataset. The data was reportedly scraped from the internet for use in training image-generation AI models, highlighting significant privacy risks associated with common data collection practices in the AI industry.

AI company Perplexity is expanding into India, viewing the country's large, digitally-savvy population as a key market to compete with rival OpenAI, as per a report in TechCrunch. The strategy targets the next phase of AI adoption, which involves transitioning artificial intelligence from early tech adopters to mainstream users. Perplexity, which operates as a conversational answer engine providing sourced information, sees India as an opportunity to gain significant user traction more quickly than in more saturated Western markets. This move highlights the growing competition among major AI firms to secure a foothold in large emerging economies.

An advisory board has recommended that OpenAI continue and strengthen its nonprofit oversight structure, arguing the technology is too consequential to be governed by a corporation alone, according to a report in AP. The recommendations were delivered in a report that also presented a broad vision for democratizing AI and reforming philanthropy. Daniel Zingale, the convener of the advisory group, stated that AI is "too important to entrust to any one sector, the private sector or even the government sector," emphasizing the need for multi-sector governance.

Netflix has confirmed the first-time use of visual effects generated by artificial intelligence in one of its original productions, as per a report in BBC. This development marks a notable implementation of generative AI in mainstream entertainment, following recent advancements in video generation technology and the 2023 Hollywood labor strikes where AI's role was a central point of negotiation. The use of such technology impacts production workflows and has implications for the future roles of visual effects artists.

Achieving scalable performance for artificial intelligence applications relies on key optimization strategies for Graphics Processing Units (GPUs), according to a report in IEEE Spectrum. The focus is on high-performance inference, the process of using a trained AI model to make predictions, where efficiency and speed are paramount. Core techniques to enhance performance include sophisticated batching, which processes multiple data inputs simultaneously, and parallelism, which distributes computational tasks across multiple GPU cores or devices. These methods are critical for maximizing throughput and reducing latency, thereby lowering the operational costs of deploying large-scale AI models and enabling more responsive real-time AI services.

Meta Platforms has hired a pair of key artificial intelligence researchers who previously worked at Apple, according to a report in Bloomberg. The move comes shortly after Meta also recruited the researchers' former manager from the iPhone maker, highlighting the intensifying competition among major technology companies to secure top-tier talent in the field of artificial intelligence.

A new survey shows AI risk is a primary challenge for cybersecurity leaders, with nearly a quarter of them reporting their company experienced an artificial intelligence-powered attack in the past year, according to a report in Bloomberg. The findings underscore growing concerns over the malicious use of AI to create more efficient and sophisticated cyberattacks, which can be used for purposes such as generating highly convincing phishing campaigns or automating the exploitation of system vulnerabilities.

The valuation for artificial intelligence startup Perplexity has surpassed $18 billion as it attracts new investors to back a higher price tag, according to a report in the Financial Times. This new valuation for the AI-powered answer engine, which provides direct, summarized responses to user queries, comes several months after its most recent funding round. The deal underscores strong investor confidence in the generative AI sector and highlights the increasing competition for traditional search engines.

A startup named Decart has developed an artificial intelligence model that can alter live video footage as it is being streamed, according to a report in Wired. The technology is designed to apply complex visual transformations and stylistic effects directly to live video feeds, moving beyond simple filters. This capability is aimed at live streaming applications, providing advanced visual tools for content creators.

Amazon is cutting an unspecified number of jobs in its Amazon Web Services (AWS) cloud-computing division, attributing the move to rising costs associated with artificial intelligence, according to a report in Bloomberg. The layoffs are part of a wider tech industry trend of reallocating resources and capital towards AI development. These high costs stem from significant capital expenditures on specialized hardware, increased energy consumption for training models, and the recruitment of specialized talent.

QTS founder Chad Williams has departed from the company, which Blackstone Inc. acquired for $10 billion four years ago, according to a report in Bloomberg. In the period following the acquisition, QTS grew to become North America's largest data center landlord, playing a significant role in providing the critical infrastructure needed to support the artificial intelligence boom. Williams's departure marks a leadership change for the firm as it continues its expansion in the competitive AI-driven market.

Swedish artificial intelligence startup Lovable, which focuses on streamlining software development, has secured $200 million in a new fundraising round, according to a report in Bloomberg. This investment boosts the company's valuation to $1.8 billion, officially making it Europe's newest 'unicorn'—a privately held startup valued at over one billion dollars.