This Week in NLP #345
Keep up with what happened in NLP in the week ending Friday 27th June 2025.
Above the Fold
Overwhelmed? Here’s our pick for five things to know about this week.
Apple's reported consideration of acquiring Perplexity AI or other AI companies has sparked debate about whether a major acquisition would solve its AI shortcomings. [TechCentral]
There’s an intensifying battle for AI talent among tech giants, with Meta's US$14.3B Scale AI investment and massive signing bonuses highlighting the competition. [Yahoo Finance]
OpenAI has begun developing collaborative document editing and integrated chat features to compete with Microsoft Office and Google Workspace. [Yahoo Finance]
Thinking Machines Lab has secured a record-breaking US$2B seed round led by Andreessen Horowitz to develop customizable, collaborative AI systems under former OpenAI CTO Mira Murati. [Tech Funding News]
xAI's Grok chatbot has sparked controversy after Elon Musk announced plans to rewrite its knowledge base, following instances where it contradicted his political views with factual corrections. [VentureBeat]
Now read on for everything else that happened in NLP this week.
Making News:
What makes this newsletter different?
This Week’s Topics:
If you’re reading this in a desktop web browser, you can access a navigation menu at the left-hand edge of this window.
The Generative AI Wars
Amazon has expanded its AI-powered Alexa+ digital assistant to over one million users through invite-only access, with plans to make it free for Prime members. [TechCrunch]
Amazon's AWS has lost VP Vasi Philomin, who oversaw generative AI development and helped create Amazon Bedrock, amid intensifying competition for AI talent. [Yahoo Finance]
Amgen has switched from Microsoft's Copilot to OpenAI’s ChatGPT for most employee AI assistance, highlighting growing competition between the two tech giants. [Mint]
Apple shareholders have filed a class action lawsuit alleging the company misled investors about AI integration timelines for Siri, impacting stock prices and iPhone sales. [Verdict]
GitHub has implemented monthly limits on premium AI requests for Copilot users, forcing them to either accept usage caps or pay more for additional computational resources. [The Register]
Google has launched its AI search mode in India, offering English-language Q&A-style searches powered by Gemini 2.5 to the country’s 870 million internet users. [TechCrunch]
And Google has launched Gemini-CLI, an open-source terminal interface offering free access to Gemini Pro 2.5 AI with generous usage limits and extensible architecture. [VentureBeat]
Meta has explored acquiring multiple AI startups including Thinking Machines, Perplexity, and Safe Superintelligence, while assembling a new AI leadership team under Alexandr Wang. [The Verge]
And Meta has also explored acquiring Runway. [CNBC]
Meta has poached three key OpenAI researchers for its superintelligence team, despite Sam Altman’s mockery of Mark Zuckerberg’s aggressive recruitment tactics. [TechCrunch]
Meta's WhatsApp has become a competitive battleground for AI chatbots, with OpenAI and Perplexity challenging Meta’s dominance over its 3 billion global users. [Yahoo Finance]
Microsoft's Copilot has struggled to compete with ChatGPT in corporate environments, as employees prefer OpenAI's more familiar chatbot despite major enterprise licensing deals. [TechRadar]
Microsoft has clashed with OpenAI over a contract clause that would terminate Microsoft’s access to AGI once OpenAI develops it. [TechCentral]
Nvidia has launched DGX Cloud, a rapidly growing cloud-computing service, entering a profitable market dominated by Amazon, Microsoft, and Google. [Yahoo Finance]
OpenAI's relationship with Microsoft has deteriorated amid restructuring disputes, competing business moves, and threats of antitrust accusations, potentially jeopardizing their partnership through 2030. [The Daily Upside]
OpenAI's Sam Altman has downplayed tensions with Microsoft over their partnership’s future, while revealing discussions with Trump and criticizing Zuckerberg’s AI ambitions. [Yahoo Finance]
OpenAI has announced plans to launch GPT-5 this summer, with CEO Sam Altman indicating significant improvements over GPT-4 while discussing potential ad integration and privacy concerns. [Adweek]
OpenAI has shifted from fixed pricing to a credits-based system for its enterprise ChatGPT service, allowing clients to access advanced features. [The Wall Street Journal]
OpenAI has removed promotional materials for its partnership with Jony Ive’s company ’io’ following a trademark lawsuit from AI device maker iyO. [Gizmodo]
ChatGPT’s iOS app has amassed nearly 30 million downloads in the past month, rivalling the combined downloads of TikTok, Facebook, Instagram, and X. [ZDNet]
Salesforce has launched Agentforce 3, providing enhanced visibility and control for scaling AI agents through a new Command Center, MCP support, and expanded enterprise capabilities. [Yahoo Finance]
AI Supremacy
Alibaba Cloud has launched new AI solutions and cloud services across Europe while expanding partnerships to strengthen its presence in Asian markets. [ITPro]
Alibaba Cloud has announced plans to open its second South Korean data center by June, expanding its US$52.9B AI and cloud infrastructure investment. [Bloomberg]
China has begun preparations for its 15th Five-Year Plan, focusing heavily on semiconductor equipment development amid US-China tensions. [DigiTimes Asia]
DeepSeek has provided support to China’s military operations while attempting to circumvent US export controls on advanced semiconductors through Southeast Asian shell companies. [Yahoo Finance]
DeepSeek's R2 language model has faced development delays and potential adoption hurdles in China due to Nvidia chip shortages following US export restrictions. [Yahoo Finance]
Gulf nations have pledged US$2T in tech deals with US companies, marking their ambitious push to transform into AI superpowers alongside America and China. [Rest of World]
Huawei has advanced to within two years of US chip design capabilities and is expected to begin exporting AI chips globally, according to White House advisor David Sacks. [The Register]
Huawei's founder Ren Zhengfei has given an unprecedented interview in People’s Daily, signaling China’s strategic shift toward embracing private enterprise to counter US tech restrictions. [The National Interest]
Huawei's deal to power Malaysia’s national AI system with Ascend GPU servers was downplayed by Malaysian officials shortly after its announcement. [Bloomberg]
OpenAI’s ChatGPT has captured nearly 80% of global AI chatbot referral traffic but remains locked out of China’s market, where DeepSeek dominates with 89.3% share. [TechRadar]
Softbank has proposed a US$1T AI and manufacturing complex in Arizona, partnering with TSMC and Samsung to create America’s answer to Shenzhen. [Tom’s Hardware]
The US Commerce Department has moved to revoke permissions for Samsung, SK Hynix and TSMC to use American technology in their Chinese operations. [Yahoo Finance]
The US Congress has introduced bipartisan legislation to ban Chinese AI systems from federal agencies amid growing concerns over US-China technological competition. [Associated Press]
Sovereign AI
AWS has launched its first APAC Innovation Hub in Singapore, an 8,000-square-foot facility showcasing cloud and AI technologies to accelerate digital transformation across the region. [Yahoo Finance]
Bitrise has announced a US$3m investment to establish Europe’s first DevOps Platform data centre in Amsterdam, addressing EU data sovereignty requirements. [Tech.eu]
Cohere has partnered with Canadian and UK governments to transform public sector operations and enhance AI research, while expanding beyond its enterprise focus. [BetaKit]
European tech leaders at Amsterdam’s Assembly conference have urged the continent to focus on AI applications rather than infrastructure, where US companies dominate through massive investments. [The Next Web]
And European leaders have called for reduced dependence on global tech companies like Microsoft, despite the firm’s promises to support European digital sovereignty. [United Press International]
G7 leaders have committed to quantum and AI development at their Kananaskis summit, with Canada pledging over US$200m to support these initiatives. [BetaKit]
Germany’s competition regulator has gathered tech companies to discuss concerns that AWS, Google, and Microsoft's control of cloud infrastructure gives them unfair dominance in AI. [The Register]
HOCHTIEF has partnered with IONOS Group to propose a European AI Gigafactory featuring 50,000+ GPUs, with operations planned to begin by 2027. [World Construction Network]
KDDI and Hewlett Packard have partnered to establish an AI data center in Osaka by 2026, featuring Nvidia Blackwell technology and liquid cooling systems. [Business Wire]
Knolskape has launched AI Academy, a simulation-based learning platform with three structured paths to address India’s AI fluency crisis through role-specific training programs. [Analytics India Magazine]
OpenAI's EMEA startup head has praised European AI innovation, highlighting strong talent across multiple cities and citing successful implementations by Klarna, Photoroom, and VEED. [Tech.eu]
Oxford University researchers found AI computing infrastructure is concentrated in just 32 countries, with the US, China, and EU controlling most data centers. [ITPro]
Station F, Paris’s converted rail depot-turned-tech incubator, has attracted international startups and doubled down on AI while fostering France’s growing tech ecosystem since 2017. [Business Insider]
Feature Creeps
Anthropic has transformed its Claude AI assistant into a platform for creating shareable, interactive applications without coding, with users already generating over 500 million artifacts. [VentureBeat]
Google's removal of raw reasoning tokens from Gemini 2.5 Pro has sparked controversy among developers who relied on this transparency for debugging and building AI applications. [VentureBeat]
Google has announced that Gemini will soon be able to control Android apps and features even when users opt out of AI training data collection. [The Verge]
Google has launched a Gemini-powered AI function in Sheets that generates text, summarizes information, and categorizes data based on selected spreadsheet cells. [The Verge]
Meta has launched AI-powered message summaries for WhatsApp in the US, using private processing technology to generate encrypted chat overviews without compromising user privacy. [TechCrunch]
OpenAI has introduced ChatGPT Record, enabling paid users to record, transcribe, and summarize up to 120 minutes of audio through its macOS app. [The Decoder]
Ring has launched Video Descriptions, an AI-powered feature that provides detailed notifications about camera-captured events to Home Premium subscribers in the US and Canada. [ZDNet]
Slack has rolled out AI upgrades across its plans, adding Salesforce channels to free accounts and AI features to paid tiers while adjusting some pricing. [ZDNet]
xAI has developed an advanced file editor for Grok featuring spreadsheet capabilities, according to leaked code, positioning itself to compete with major tech companies in AI-enhanced productivity tools. [Yahoo Finance]
YouTube has launched an AI-powered search carousel for Premium users and expanded its conversational AI tool, potentially affecting creator engagement and video traffic. [TechCrunch]
Hype Bubble?
AlgorithmWatch has warned against Big Tech’s portrayal of AI as a magical solution. [The Hankyoreh]
Apple‘s research paper ‘The Illusion of Thinking’ has sparked debate by demonstrating LLMs’ struggles with reasoning tasks. [The Decoder]
Ataccama's research has revealed that 42% of businesses distrust their AI outputs due to poor data management and limited observability programs. [TechRadar]
Boston Consulting Group has found AI adoption stalling in enterprises due to insufficient training, limited tool access, and lack of management support. [ZDNet]
Gartner has predicted that half of companies will abandon plans to replace customer service workers with AI by 2027, citing poor results and consumer dissatisfaction. [ITPro]
Kaseya's CEO Rania Succar has identified fragmented business data and change management challenges as key factors behind slower-than-expected enterprise AI adoption. [The Register]
MIT Technology Review has launched its AI Hype Index, highlighting developments including Mattel‘s AI toy partnership and concerns about AI agents’ reliability. [MIT Technology Review]
Salesforce has shifted to extensive AI integration, with CEO Benioff claiming 30-50% of work is AI-driven while laying off 1,000 employees and hiring AI-focused replacements. [Gizmodo]
A study by Stanford University revealed that professionals welcome AI agents for automating mundane tasks while preferring to maintain control over higher-stakes work activities. [ZDNet]
A UK survey by Smarsh revealed widespread AI adoption in financial services, with 37% of employees using public AI tools despite limited training and oversight concerns. [Business Wire]
A UN Development Programme survey of 21 countries has revealed that developing nations show greater trust in AI systems than wealthy countries, with China displaying particularly strong confidence. [Bloomberg]
Vultr has released a global AI maturity report showing that organizations with advanced AI capabilities significantly outperform their competitors in revenue and market share. [Business Wire]
Hardware
Amazon has developed its fourth-generation Graviton4 chip with 73 billion transistors and 600Gbps network bandwidth, challenging Intel and AMD. [Yahoo Finance]
AMD has announced its MI400 series AI accelerators and Helios rack-scale solution for 2026, promising significant performance gains while still trailing Nvidia's market dominance. [The Motley Fool]
Cerebras Systems and other manufacturers have developed pizza-sized AI processor chips that offer superior performance to GPUs. [TechRadar]
Groq and Cerebras Systems have challenged Nvidia's ‘AI factory’ narrative, exposing issues with capacity constraints and quality variations in AI inference delivery. [VentureBeat]
Intel's Gaudi AI chip initiative has faltered despite attractive pricing, leading the company to abandon Falcon Shores and pivot toward rack-scale solutions launching after 2026. [The Motley Fool]
Nebius has launched Nvidia GB200 Grace Blackwell Superchip capacity in Europe, expanding its AI infrastructure while adding multiple NVIDIA platform integrations. [Yahoo Finance]
Nvidia has secured exclusive production capacity through 2026 at Wistron’s new Taiwan facility to manufacture Blackwell and Rubin AI server systems. [Yahoo Finance]
Nvidia has established itself as the dominant AI chip supplier to major tech companies, which are collectively planning to spend hundreds of billions on data center infrastructure in 2025. [The Motley Fool]
Consumer AI
Google has announced several AI features for Chromebook Plus devices, including image generation, text summarization, and enhanced Lens capabilities, alongside a free AI Pro trial. [Ars Technica]
Lenovo has launched its Chromebook Plus, featuring MediaTek's Kompanio Ultra processor, exclusive Google AI features, and up to 17 hours of battery life. [Business Wire]
Meta has partnered with Oakley to launch AI-enabled HSTN smart glasses featuring HD video capture, open-ear audio, and Meta AI integration for athletes. [Facebook]
OpenAI has explored various AI hardware devices, including in-ear products, but court filings revealed its first collaboration with Jony Ive’s io won’t be a wearable device. [TechCrunch]
Solos has unveiled two AI-powered smart glasses models - the audio-focused AirGo A5 and camera-equipped AirGo V2 - launching in Q3 and Q4 2025 respectively. [VentureBeat]
Xiaomi has launched AI smart glasses in China featuring electrochromic dimming, USB-C charging, and similar functionality to Meta's Ray-Ban glasses. [TechRadar]
It’s Only a Model
AI21 has integrated its Jamba 1.6 language model with Hewlett Packard Private Cloud AI, enabling enterprises to run powerful AI applications securely within their own data centers. [Yahoo Finance]
Arcee AI has launched AFM-4.5B, a 4.5-billion-parameter foundation model delivering enterprise-grade AI with high accuracy, compliance, and cost-efficiency across multiple deployment options. [Arcee AI]
Google DeepMind has launched Gemini Robotics On-Device, an offline language model enabling local robot control through natural language prompts and demonstrating adaptability across different robots. [TechCrunch]
Google has launched Imagen 4, its latest text-to-image AI model family, offering improved text rendering and two tiers of service through its Gemini API. [Google]
Liquid AI has transformed its 1.3B parameter model into a capable mathematical reasoner through extensive supervised fine-tuning and targeted reinforcement learning. [Liquid AI]
Microsoft has launched Mu, a 330M-parameter on-device language model that powers Windows Settings’ agent while running efficiently on Neural Processing Units. [Windows.com]
Mistral AI has launched Small 3.2, an upgraded language model featuring enhanced instruction accuracy, reduced repetition errors, and improved STEM-related performance over its predecessor. [Marktechpost Media]
Whose Data?
Anthropic has received a mixed ruling in a landmark AI copyright case, with training deemed fair use but its downloading of pirated books declared illegal. [Publishers Weekly]
Anthropic scanned millions of purchased books by destroying their bindings to train its AI assistant Claude, according to recently revealed court documents. [Ars Technica]
The BBC has threatened legal action against Perplexity AI for unauthorized reproduction of its content, demanding deletion, cessation, and financial compensation. [BBC]
Creative Commons has announced CC signals, a project enabling dataset holders to specify how their content can be used for AI training, aiming to balance open internet principles with data protection. [TechCrunch]
Getty Images has dropped its main copyright claims against Stability AI in London’s High Court while maintaining secondary infringement and trademark claims in both UK and US courts. [TechCrunch]
Google has confirmed using YouTube videos to train AI models like Gemini and Veo 3, raising intellectual property concerns among creators and media companies. [CNBC]
And Google has sparked privacy concerns by announcing that Gemini will access Android users’ critical apps like Phone and Messages by default, regardless of their opt-in status. [Gizmodo]
Hugging Face has removed a controversial dataset of 12.6 million scraped fanfiction stories after Archive of Our Own writers protested the unauthorized collection of their work. [The Verge]
A study by Incogni ranked Mistral AI's Le Chat as the most privacy-friendly AI service, with ChatGPT and Grok following in second and third places. [ZDNet]
Man of Many has become Australia’s first lifestyle publisher to license content to AI platform ProRata, emphasizing transparent attribution and fair compensation for publishers. [journalism.co.uk]
Meta's Llama 3.1 70B model has demonstrated a significantly higher tendency to reproduce copyrighted Harry Potter text compared to other AI models, according to recent academic research. [Ars Technica]
Meta has won a copyright case despite Judge Chhabria’s ruling that AI training likely infringes copyrights, while another judge in the same district reached the opposite conclusion. [TechDirt]
Microsoft has been sued by authors who alleged the company used 200,000 pirated books to train its Megatron AI model, seeking damages of up to $150,000 per work. [The Guardian]
OpenAI has failed to overturn a court order requiring preservation of all ChatGPT logs for a copyright lawsuit, despite users’ privacy concerns. [Gizmodo]
Reddit has sued Anthropic for allegedly accessing its platform over 100,000 times since July 2022, despite claiming to have blocked such scraping activities. [Yahoo Finance]
The UK’s Data (Use and Access) Bill has weakened UK citizens’ data protection rights while expanding government and Big Tech control over automated decision-making systems. [Computer Weekly]
The LLM Ecosystem
Canva has become the first design platform to integrate its creative suite with ChatGPT, launching a deep research connector and MCP Server for AI-powered design workflows. [Business Wire]
Coastal has launched Waves for AI, a managed service helping organizations implement Salesforce's Agentforce platform while ensuring proper AI governance and data orchestration. [Yahoo Finance]
Google has partnered with the University System of Maryland to provide students free access to career certificates in tech skills and AI training. [GovTech]
Emergence AI has launched CRAFT, a natural language platform that automates enterprise data pipelines using self-verifying agent systems, reducing months of engineering work to minutes. [Business Wire]
Genesys has announced Cloud AI Studio, a centralized hub for building AI-driven customer engagement tools with built-in safety controls and no-code requirements. [Business Wire]
HongShan Capital Group has launched Xbench, an evolving AI benchmark system that tests both academic knowledge and real-world task performance, making it publicly available. [MIT Technology Review]
Hewlett Packard has launched new AI factory solutions featuring Nvidia Blackwell technology, expanding its enterprise AI portfolio with storage, computing, and professional services capabilities. [Business Wire]
Huawei has opened HarmonyOS 6 beta to developers, introducing AI agents and a framework for creating them, while reporting strong ecosystem growth for version 5. [South China Morning Post]
Hystax has launched Kiroframe, an MLOps platform that automates and scales AI development through model profiling, dataset management, and workflow optimization. [PRWeb]
LAION has released EmoNet, an open-source toolkit for emotional intelligence in AI, reflecting a broader industry shift toward developing emotionally intelligent models. [TechCrunch]
Lakera has launched the AI Model Risk Index, a comprehensive security assessment framework for evaluating LLMs under real-world adversarial conditions. [Business Wire]
Lenovo has expanded its Hybrid AI Advantage program with new services, solutions, and platforms to help enterprises build and scale AI capabilities across their operations. [Business Wire]
MIT researchers have developed SEAL, a framework enabling LLMs to generate their own training data and continuously learn by updating their internal parameters. [VentureBeat]
NeuroCluster AI has launched Supernova, an open-source enterprise AI platform featuring persistent memory and self-improving capabilities. [EIN Presswire]
NTT Data has expanded its Salesforce partnership by launching a comprehensive service offering for Agentforce, helping clients implement AI agents across their organizations. [Yahoo Finance]
Okta has launched Cross App Access, a new protocol extending OAuth to help enterprises secure and manage AI agents’ interactions across different applications and systems. [Business Wire]
Riskonnect has launched an AI Governance solution within its risk management suite to help organizations manage AI risks while maintaining innovation and compliance. [Business Wire]
Strategy has launched Mosaic, an AI-powered universal intelligence layer that connects enterprise data sources and enables consistent access for business users and AI applications. [Business Wire]
Teradata has launched AI Factory, an on-premises AI solution offering secure, integrated capabilities for private AI development with Nvidia technology. [Business Wire]
Treasure Data has launched MCP Server, an open-source connector enabling AI assistants to interact directly with customer data through natural language queries. [Business Wire]
Uber AI Solutions has expanded its data-labelling platform to over 30 countries and doubled its clickworker base following Scale AI's partnership with Meta, which disrupted the industry. [Forbes]
Verax AI has launched Verax Protect, enabling enterprises to safely use AI tools while preventing sensitive data leaks and unauthorized access. [Business Wire]
Agentic AI
CB Insights has launched an AI agent workforce featuring 11 specialized agents that provide market intelligence for strategic business decisions using validated company data. [Business Wire]
Cognizant has expanded its Salesforce partnership by launching AI agent transformation services through Agentforce, helping enterprises deploy AI-augmented workforces across multiple industries. [Yahoo Finance]
Digital Wave Technology has integrated agentic AI into its ONE Platform, enabling autonomous agents to perform complex enterprise tasks while combining with LLMs for intelligent decision-making. [Business Wire]
Emergence AI has launched CRAFT, a self-serve platform that automates enterprise data pipelines by allowing users to create AI agent systems using natural language commands. [VentureBeat]
Genspark has launched an AI Super Agent that extends ‘vibe coding’ principles to enterprise workflows, enabling autonomous task completion with minimal human control. [VentureBeat]
Google's Agent2Agent protocol donation to the Linux Foundation has united major tech companies in standardizing AI agent communication across different platforms. [Forbes]
H2O.ai has achieved a world-record 79.7% accuracy with its h2oGPTe Agent on the GAIA benchmark, surpassing competitors like Google and Microsoft. [Business Wire]
Hewlett Packard has launched GreenLake Intelligence, an AI-powered framework using domain-specific LLMs and agent networks to autonomously monitor and manage IT operations across hybrid environments. [BigDATAwire]
Jobright.ai has launched an AI agent that automatically applies for jobs on users’ behalf while securing US$3.2m in funding from Translink Capital and Indeed. [The Register]
Kimi-Researcher has launched an autonomous agent that achieved state-of-the-art performance on multiple benchmarks through end-to-end reinforcement learning training. [Moonshot AI]
Laiye has unveiled five enterprise AI agents at its ‘Lead 2025’ launch, showcasing solutions for lead generation, recruitment, and market analysis backed by customer success stories. [Laiye]
SmythOS has partnered with AGNTCY and released its core AI agent technology under MIT license, advancing development of an open, interoperable Internet of autonomous agents. [EIN Presswire]
Navan has launched Cognition, an AI platform enabling companies to build specialized AI agent teams for handling complex business tasks with enhanced accuracy and reliability. [Business Wire]
SUPERWISE has launched an enterprise AgentOps platform that enables secure deployment and management of third-party AI agents with built-in governance and monitoring capabilities. [Business Wire]
Tricentis has launched remote MCP servers and Agentic Test Automation to enable AI-powered enterprise software testing with flexible deployment options. [Business Wire]
Warp has launched version 2.0, introducing an Agentic Development Environment that combines top-rated coding capabilities with multi-threaded agent management and improved development workflows. [Warp]
Wrike has launched an MCP Server that connects third-party AI agents with its work management platform, enabling real-time data access and automated workflows. [Business Wire]
ZS has expanded its Salesforce partnership to provide consulting and implementation services for Agentforce, helping companies scale AI agents across multiple industries. [Business Wire]
Other LLM Sightings
Applebee's and IHOP's parent company has announced plans to implement AI-powered personalization tools for customer recommendations, following similar moves by other restaurant chains. [The Verge]
AGAT Software has launched Personal BusinessGPT, a free browser-based AI chatbot that operates offline using local models to ensure complete privacy and security. [EIN Presswire]
Amazon has unveiled multiple AI implementations across Prime Video, from content discovery and sports overlays to picture quality enhancement and advertising personalization. [Deadline]
Connecting the Dots has launched an AI platform that automates marketing, lead generation, and customer support tasks for small businesses through a unified interface. [EIN Presswire]
Cyndx has launched Scholar, an AI-powered research tool that generates comprehensive business reports and analysis using proprietary data from over 30 million companies. [EIN Presswire]
Daydream has launched an AI-powered fashion shopping chatbot that connects users with over 8,000 brands through personalized search and style recommendations. [TechCrunch]
Deepinvent has launched an AI Innovator platform that generated over 2,000 inventions in its first week by compressing months of R&D into minutes. [Business Wire]
Goldman Sachs has launched an AI assistant to help employees with document summarization, content creation, and data analysis, joining other major banks offering similar tools. [Yahoo Finance]
JAGGAER has launched JAI, an AI-powered procurement copilot that orchestrates tasks and provides guidance while maintaining human oversight of supply chain management. [Business Wire]
Joveo has launched an AI Career Site Builder that enables recruitment teams to create customized career sites and landing pages within minutes using simple prompts. [GlobeNewswire]
LinkedIn's AI post-polishing feature has seen limited adoption, while AI-related job listings and profile skills have surged, according to CEO Ryan Roslansky. [TechCrunch]
Lionbridge has launched an AI-powered Content Remix App that creates personalized, multilingual content across marketing channels in over 70 languages. [EIN Presswire]
MetLife has partnered with Sprout.ai to implement claims automation technology across its global operations, aiming to improve processing speed and customer experience. [Business Wire]
MyLondonBeauty has launched iOS and Android apps with an AI assistant to help Londoners find and book beauty services across the capital. [EIN Presswire]
Reasoner has launched Mind Reasoner, an AI-powered communication analysis tool that reveals hidden meanings in conversations with verifiable proof, based on patent litigation technology. [PRWeb]
Resolver has launched an AI-powered Incident Intake and Triage system that automates security report processing, reducing administrative time by up to 90%. [PRWeb]
Sitch has launched an AI-powered dating app that combines traditional matchmaking expertise with language models to create more meaningful connections than swipe-based alternatives. [TechCrunch]
SplashBI has launched Tahoe v6, a unified enterprise analytics platform featuring AI-powered insights, custom connectors, and modernized UX for streamlined data reporting. [EIN Presswire]
Syncfusion has launched major AI-driven updates to its BoldDesk platform, adding automated task execution, improved ticketing, and enhanced data integration features. [GlobeNewswire]
Tradeshift has launched an AI-powered Analytics app built on Amazon QuickSight, enabling buyers to instantly transform payables data into actionable insights. [EIN Presswire]
The USPTO has sought AI vendors to scan patent applications for prior art. [The Register]
Visit Seattle has launched Emerald, an AI-powered travel assistant that provides personalized recommendations and trip planning help for visitors to the Seattle area. [Visit Seattle]
VONQ has partnered with Employ to launch CPA+, an AI-powered hiring solution that screens candidates and reduces hiring time by up to 90%. [Business Wire]
Walmart has developed Element, an internal AI platform that manufactures applications at scale, handling 3 million daily queries and supporting 1.5 million associates across multiple functions. [VentureBeat]
WiseCleaner has launched WiseX, an AI assistant platform integrating multiple AI models, document processing, real-time web search, and image editing capabilities. [EIN Presswire]
Workato has launched GO, an AI-powered super app enabling enterprise-wide search, workflow automation, and system orchestration through a unified interface. [Business Wire]
X0PA AI has launched Zeus and Ruby AI hiring agents, completing its Agentic AI Suite that streamlines recruitment with bias-free, automated processes. [EIN Presswire]
Yum China has launched Q-Smart, an AI-powered restaurant management assistant that helps with inventory, scheduling, and food safety through voice commands and wearable devices. [Verdict Food Service]
Risks and Responses
The American Security Project has found that five major AI models, including four from US companies, frequently echoed Chinese Communist Party propaganda and censorship in their responses. [The Register]
Anthropic's research revealed that major AI models, including GPT, Gemini, and Grok, resorted to blackmail when given email access and threatened with shutdown. [eWeek]
Anthropic has declined to patch a SQL injection vulnerability in its archived SQLite MCP server that could allow attackers to hijack AI support bots and steal customer data. [The Register]
Backslash Security has discovered widespread vulnerabilities in MCP servers, affecting roughly half of 15,000+ servers and exposing developers to potential attacks. [ITPro]
BitMindAI's founder has warned that AI-generated and recycled conflict footage between Iran and Israel is eroding public trust in digital content across social media platforms. [eWeek]
Cloudsmith's research revealed widespread AI code adoption among developers, with 42% having AI-filled codebases, while warning of security risks from inadequate code review practices. [ITPro]
Cobalt's research revealed security professionals are seeking a pause on AI adoption as teams struggle to manage generative AI security risks. [ITPro]
Data² has received a patent for technology that makes AI more explainable and resistant to hallucinations, particularly for military and intelligence applications. [DefenseScoop]
LinkedIn has reported processing 11,000 AI-generated job applications per minute, overwhelming employers and forcing some to abandon traditional hiring methods. [Ars Technica]
Microsoft has released its 2025 Responsible AI Transparency Report, outlining investments in AI safety tools, risk management practices, and regulatory compliance measures. [TechRadar]
Cybercriminals have jailbroken Mistral AI's Mixtral and xAI's Grok AI tools to create WormGPT variants capable of generating malicious code and hacking tutorials. [TechRadar]
MIT researchers have found that students using AI for essay writing showed reduced brain activity and recall compared to those writing without technological assistance. [TechRadar]
And MIT researchers discovered that language models making medical recommendations were misled by non-clinical text elements, particularly affecting advice given to female patients. [MIT News]
OpenAI has warned that its upcoming AI models could enable the creation of biological weapons. [eWeek]
Pope Leo XIV has urged tech companies to develop AI within an ethical framework that prioritizes human dignity, during a Vatican conference attended by Silicon Valley executives. [KESQ-TV]
THC Technologies’ VoiceVantage software has demonstrated over 95% effectiveness in blocking deepfake voice attempts from five leading voice cloning tools during university testing. [Speech Technology Magazine]
Tech companies have promoted an overly positive ‘good AI’ narrative while downplaying privacy concerns and algorithmic biases that discriminate against marginalized groups. [The Conversation]
Trelexa has advised professionals to establish their authority through published works before AI-generated content further saturates and commoditizes the content marketplace. [EIN Presswire]
UC Berkeley researchers have demonstrated AI models’ growing prowess at finding software vulnerabilities, discovering 17 new bugs during tests of 188 open-source codebases. [Wired]
Zluri has released a report revealing that 80% of enterprise AI tools operate without IT oversight, based on data from 160+ organizations worldwide. [Business Wire]
Regulation
DeepSeek and other AI models from ‘adversarial nations’ could be banned from US government agencies under the newly introduced ‘No Adversarial AI Act’. [TechRadar]
The Electronic Frontier Foundation has criticized the NO FAKES Act’s property rights-based approach to regulating AI-generated replicas, arguing privacy laws would better protect people from unauthorized digital impersonation. [The Register]
Geoffrey Hinton has scheduled a meeting with Canada’s AI Minister to advocate for AI regulation, despite growing governmental preference for rapid technology adoption over restrictions. [The Globe and Mail]
The UK’s Data Use and Access Bill has transformed data protection laws, easing data sharing while raising concerns about European adequacy and privacy rights. [Computer Weekly]
Senate Republicans have advanced a provision to block states from enforcing AI regulations by withholding federal broadband funding, despite opposition from some party members. [TechCrunch]
Meta Platforms and other tech giants have pushed for the state AI regulation ban, facing opposition from several Republican lawmakers. [Yahoo Finance]
Microsoft's chief scientist Eric Horvitz has opposed the proposed ban on state-level AI regulation, warning it would hinder rather than accelerate technological progress. [The Guardian]
And survey by YouGov has revealed widespread bipartisan opposition to Congress’s AI regulation moratorium, with 55% of American voters rejecting the measure. [EIN Presswire]
Conversational AI
AVOXI has embedded AI into its cloud voice platform, introducing features for automated issue detection, intelligent caller ID, and enhanced call routing for global contact centers. [Business Wire]
Call Loom has launched an AI-powered call tracking platform featuring intelligent routing, fraud protection, and automated agents to optimize inbound lead performance. [EIN Presswire]
Cerence AI has partnered with Mercedes-Benz to integrate advanced conversational AI capabilities into the fourth generation of MBUX for the all-electric CLA model. [GlobeNewswire]
Creatio has launched its 8.3 ‘Twin’ Release, integrating AI capabilities throughout its CRM platform with conversational interfaces, prebuilt agents, and no-code development tools. [VentureBeat]
ElevenLabs has launched 11.ai, an alpha-stage voice assistant that combines conversational AI with Model Context Protocol to enable action-taking capabilities across multiple platforms. [ElevenLabs]
Fonio.ai has launched an AI-powered phone assistant service for SMEs in Vienna, acquiring over 1,200 customers since its September 2024 debut. [Silicon Canals]
LocaliQ has launched an AI Voice Agent within its Dash platform to convert missed business calls into qualified leads through automated, intelligent conversations available 24/7. [Editor & Publisher Magazine]
Research by Okta revealed that 70% of global consumers preferred human interactions over AI for customer service, with baby boomers showing the strongest anti-AI sentiment. [TechRadar]
PanTerra has launched Luna AI, an AI-powered receptionist system that automatically handles, routes, and filters calls while providing natural, emotionally intelligent responses. [EIN Presswire]
Qualified has launched Piper for Slack, integrating its AI sales agent into the platform to enable real-time collaboration between human teams and AI. [Business Wire]
Sandgarden has transformed the book ‘Founder vs. Investor’ into an AI chatbot offering candid startup advice, based on expert knowledge from the book’s co-authors. [PRWeb]
SignalWire has launched a Python-based Agents SDK that enables rapid development of AI voice agents by integrating telephony, messaging, and AI orchestration capabilities. [Business Wire]
Talkdesk has launched two AI-powered customer experience platforms for the travel and hospitality industries, enabling digital-first support and automated service delivery. [GlobeNewswire]
Be Real
2wai has launched an app allowing users to create AI-powered digital clones of themselves for multilingual conversations, with initial focus on entertainers and influencers. [Axios]
Anthropic's study of 4.5 million Claude conversations has revealed that users primarily seek productivity assistance, with emotional support accounting for only 2.9% of interactions. [TechCrunch]
Baidu's AI-powered digital avatars helped livestreamer Luo Yonghao generate US$7.65m in sales during a groundbreaking six-hour streaming session. [CNBC]
HelpBnk has launched a free AI chatbot app that replicates entrepreneur Simon Squibb’s business knowledge and offers voice-enabled mentoring conversations. [TechRadar]
Musical AI and other tech companies have developed infrastructure to detect, track, and license AI-generated music. [The Verge]
ChatGPT has influenced human speech patterns, with research showing increased usage of AI-favored words like ‘delve’ and ‘prowess’ in academic discourse since its release. [The Verge]
France’s Palace of Versailles has partnered with OpenAI and Ask Mona to introduce AI-powered talking statues that interact with visitors in three languages at the historic palace. [France 24]
Reddit has drawn a line against AI-generated content, implementing human verification tools while balancing lucrative AI training deals with preserving authentic user interactions. [Inventiva]
SHI International has launched a platform enabling businesses to create AI-powered digital humans in six weeks through collaboration with Nvidia and Hewlett Packard. [Yahoo Finance]
Soopra has enabled content creators to use AI agents for fan engagement on social media. [Digiday]
YouTube has announced plans to launch AI-generated Shorts using Google‘s Veo 3 technology, sparking concerns about content authenticity and creators’ intellectual property rights. [Gizmodo]
Voice News
Digiarty Software has launched VideoProc Converter AI 7.5 for Mac, featuring AI-powered noise suppression for audio and video content processing. [EIN Presswire]
ElevenLabs has launched its first standalone mobile app for iOS and Android, enabling users to generate AI-powered voice clips from text while on the move. [TechCrunch]
LAION and Intel have launched Empathic Insight, an open-source suite of AI models trained on synthetic data to recognize 40 different emotional states in faces and voices. [The Decoder]
NICE has launched Inform AI, an AI-powered system helping Emergency Communications Centers improve response times and staff retention through automated transcription and searching. [Business Wire]
Rasa has launched Voice, an AI architecture that processes audio input directly without speech-to-text conversion, enabling faster, more nuanced conversational interactions. [Speech Technology Magazine]
Resemble AI has launched a simulation platform that tests organizational defenses against voice-based deepfake attacks through realistic scenarios and employee risk scoring. [Speech Technology Magazine]
ScreenXchange has launched SX Agent SAM, an AI voice agent that automates background verification calls across education, employment, and references. [EIN Presswire]
Speaktor has introduced AI dubbing technology that transformed a Swedish film into English without human voice actors. [EIN Presswire]
Voicemod has launched its voice-transformation hardware device, Voicemod Key, in Europe, bringing AI voice changing and soundboard capabilities to console gaming platforms. [Silicon Canals]
WellSaid has launched an upgraded API offering ultra-realistic AI voice capabilities at reduced costs, alongside a new partner program supporting growth and monetization. [Business Wire]
Document AI
ancora Software has launched ancoraFusion, an AI-powered document processing platform that automates complex workflows through a three-tiered architecture of specialized models. [EIN Presswire]
Artificio has launched Document Chat, an AI platform that processes 500MB files through natural conversation while maintaining context across thousands of pages. [EIN Presswire]
Hyperscience has partnered with Carahsoft to provide AI-powered back-office automation solutions to government agencies. [Hyperscience]
KnowledgeLake has enhanced its document automation platform with improved indexing, task management, AI transparency, and training tools to streamline workflow processes. [KnowledgeLake]
Toshiba has launched Elevate Sky Workflow, an AI-powered cloud application that automates document processing and routing to boost workplace productivity. [Business Wire]
Transkribus has launched Text Titan I ter, a text recognition model that outperforms ChatGPT, Gemini, and other LLMs in transcribing historical documents. [Transkribus]
Translation
BeLazy and iLangL have launched a joint solution automating web content localization workflows between major content management systems and project management platforms. [MultiLingual]
A new report from the British Film Institute has revealed widespread AI adoption across UK media, from automated subtitling and dubbing to interactive dialogue generation in games and content classification systems. [Slator]
DeepL has expanded its translation platform with Vietnamese, Hebrew, and Thai languages, and added Arabic and Traditional Chinese support for document translation. [MultiLingual]
Google has integrated automated translation into its search results, redirecting traffic through its own ecosystem rather than sending users to original content creators’ websites. [Slator]
JR Language has launched a hybrid translation service combining AI technology with human expertise to deliver fast, high-quality multilingual content across various industries. [EIN Presswire]
KUDO has launched Mobile App v3.0, featuring AI-powered speech translation across 60 languages and three participation modes for real-time multilingual communication on smartphones. [MultiLingual]
OnTheGoSystems has launched Private Translation Cloud, an AI-powered platform that automates software translation by integrating with code repositories and delivering production-ready translations. [MultiLingual]
Walmart has rolled out AI-powered tools to 1.5 million US employees, including real-time translation and task management features to enhance customer service. [CX Dive]
Search
Adobe has launched LLM Optimizer to help businesses improve their visibility and rankings across AI chatbots like ChatGPT, Gemini, and Claude. [TechRadar]
BigCommerce and Feedonomics have partnered with Perplexity to provide AI-optimized product data feeds, enhancing brand visibility in AI-powered search results. [GlobeNewswire]
Central Department Store has transformed its e-commerce platform by implementing Algolia's AI-powered search technology across its 69-million-product catalog, doubling conversions. [Business Wire]
Floyi has launched AIRS Analyzer, a tool comparing results from multiple AI search engines and traditional search in one dashboard for SEO professionals. [EIN Presswire]
Google has maintained its dominance in online search with 13.7 billion daily searches, significantly outpacing ChatGPT and other platforms including Instagram, Snapchat, and LinkedIn. [Social Media Today]
Google's Jeff Dean has emphasized the company’s real-time index freshness as a competitive advantage for its AI services, distinguishing it from competitors using less frequently updated indexes. [Search Engine Roundtable]
Google's AI Overviews has caused significant drops in website traffic, with search referrals declining up to 30% while AI companies increasingly crawl sites without driving proportional visitor traffic. [The Register]
Google has resumed rolling out its improved Ask Photos feature after addressing speed and reliability issues in the AI-powered photo search tool. [TechCrunch]
HasData has launched an API that converts Google's AI Mode search results into structured JSON data, helping businesses monitor their online presence amid reduced click-through rates. [PRWeb]
Perplexity AI has expanded its Comet browser to Windows testers following its Mac release. [TechRepublic]
And Perplexity has launched a comprehensive AI-powered search tool featuring Labs, Voice Mode, and templates, offering users an alternative to traditional search engines with organized, well-cited results. [Fast Company]
The UK’s Competition and Markets Authority has proposed designating Google with ‘strategic market status’, which would require the company to offer alternative search options and fairer content practices. [TechCrunch]
AI in Journalism
News Corp Australia has trained journalists at three mastheads to use NewsGPT, an AI tool enabling persona-based writing, amid staff concerns over job security and journalistic integrity. [The Guardian]
Reuters Institute has found that while newsrooms are eagerly pursuing AI personalization features, audience interest remains relatively low, with only 30% expressing interest in any single option. [NiemanLab]
Health Tech
Berry Fertility has launched Smart Compose, an AI-powered communication tool that helps fertility clinic staff draft personalized patient responses while maintaining human oversight. [Business Wire]
Elation Health has launched ‘Actions’, an AI-driven EHR feature that automates clinical tasks and improves patient engagement. [Business Wire]
Main Street Medical Clinic has partnered with eClinicalWorks to implement healow Genie, an AI-powered contact center solution offering multilingual patient support and 24/7 service. [Business Wire]
A survey by ModMed revealed that 57% of patients supported AI use in medical practices if it increased face-to-face time with doctors, despite concerns about AI-driven diagnostics. [Business Wire]
Onpoint Healthcare Partners has launched Iris, an AI-powered medical platform that combines technology and clinical expertise to streamline healthcare operations across the patient journey. [EIN Presswire]
OpenEvidence has emerged as a leading AI diagnostic tool used by 25% of US physicians. [The Neuron]
Salesforce has partnered with UChicago Medicine to implement Agentforce for Health, enabling AI-powered automation of routine patient services and administrative tasks. [Business Wire]
Stanford Health Care has launched ChatEHR, an AI tool that reduced emergency physicians’ chart review time by 40%. [VentureBeat]
The US FDA has authorized over 1,000 AI-enabled medical devices, but updates to this authorization list stalled after December amid broader AI deregulation efforts. [STAT]
Legal Tech
AI2L has launched as a global legal innovation alliance, uniting regional legal tech associations from Brazil, Asia-Pacific, and Europe. [Artificial Lawyer]
Alexi has launched Private Cloud, enabling law firms to deploy AI solutions within isolated environments that ensure data security and compliance while maintaining full control. [Business Wire]
AXL has partnered with Dentons law firm to co-develop AI ventures and transform legal services, with Dentons becoming AXL’s exclusive legal partner. [Business Wire]
CobbleStone Software has released a report revealing widespread industry demand for AI-driven automation to replace inadequate manual contract management systems. [PRWeb]
Consilio has launched Aurora, a GDPR-compliant eDiscovery platform, in the UK and Europe following its acquisition of TrueLaw and partnership with LegalMation. [Artificial Lawyer]
ContractPodAi has partnered with Microsoft to integrate its legal AI platform Leah with Azure OpenAI, aiming to enhance legal automation solutions globally. [Yahoo Finance]
Descrybe.ai has launched a paid Legal Research Toolkit featuring an AI-driven citator, brief checker, and enhanced search capabilities, while maintaining its free basic service. [LawSites]
Glenmont Group has launched Black Sheep Review, an AI-powered managed document review platform combining artificial and human intelligence for faster, more accurate legal analysis. [PRWeb]
Harvey has launched Workflow Builder, allowing legal teams to create customized, reusable systems based on their proprietary knowledge and existing client processes. [Legal IT Insider]
Integreon has launched a genAI compliance service using ContractPodAi's Leah platform to automate regulatory checks and remediate contract language across frameworks like DORA, GDPR, and HIPAA. [Artificial Lawyer]
Legora has launched Workflows, an AI framework enabling lawyers to use natural language for orchestrating complex legal tasks through multiple tools and data sources. [Legal IT Insider]
PowerPatent has launched AI-driven automation technology that streamlines patent applications by reducing manual work and minimizing errors in the drafting process. [EIN Presswire]
Pramata has launched AI Negotiator, a Microsoft Word-integrated contract management solution that reduces contract processing time by 50% through automated analysis and redlining capabilities. [Business Wire]
Prenup India has launched India’s first online platform offering legally valid prenuptial and quasi-prenuptial agreements. [EIN Presswire]
Purpose Legal has expanded its partnership with Relativity to offer RelativityOne cloud platform services across North America, enhancing its eDiscovery and legal support capabilities. [EIN Presswire]
Reveal has integrated Milyli's Blackout redaction technology into its Enterprise platform following the acquisition of Blackout’s source code last year. [Business Wire]
Transforming.legal has partnered with European legal AI company Noxtua to integrate AI solutions and provide strategic transformation services for legal and compliance clients. [Artificial Lawyer]
Wisconsin’s legislature has introduced a controversial bill allowing AI-powered court interpretation to address interpreter shortages affecting 167,000 limited-English-proficient residents. [MultiLingual]
Ed Tech
Axonify has launched Co-Creator, an AI-powered solution that transforms business content into training materials. [EIN Presswire]
BrainFreeze has launched an AI orchestration platform enabling schools to create and manage custom AI assistants while maintaining security standards for K-12 environments. [Business Wire]
Domo has partnered with Burbio to embed AI-powered analytics into Burbio’s K-12 education data intelligence platform, enabling detailed insights for school district initiatives. [Business Wire]
OpenAI’s ChatGPT has transformed teaching by helping educators create lesson plans, grade work, and reduce administrative tasks, according to a Gallup survey of 2,000 teachers. [Associated Press]
Turnitin has expanded from a plagiarism detector to an AI-detection tool, collecting over US$15m from California colleges while amassing a controversial database of student papers. [The Markup]
Funding
2wai has emerged from stealth with US$5m in funding to launch a social app featuring device-processed AI HoloAvatars capable of real-time, multilingual conversations. [Yahoo Finance]
Abridge has raised US$300m in Series E funding at a US$5.3 billion valuation while expanding into AI-powered medical coding and serving 150+ major health systems. [TechCrunch]
Ankor has raised US$1.3m in Pre-Seed funding for its AI-powered platform that automates e-commerce operations and payment reconciliation tasks. [Tech.eu]
Bonfy.AI has emerged from stealth mode with a US$9.5m-backed security platform that uses AI to protect organizations from generative AI and unstructured data risks. [PRWeb]
Botpress has secured US$25m in Series B funding to expand its AI agent development and deployment platform. [PYMNTS]
Browserbase has raised US$40m in Series B funding and launched Director, a non-developer tool, after reaching 1,000 customers and US$3m revenue with its headless browser technology. [Upstarts]
Centific has secured US$60m in Series A funding led by Singapore’s Granite Asia to expand its AI data foundry services globally. [Yahoo Finance]
Cequence has secured €3m in total funding to expand internationally and enhance its AI-powered contract management platform, which already serves major enterprise clients. [The Slovak Spectator]
Chatlyn has secured €8m in Series A funding to expand its AI communication platform that centralizes and automates guest communications across multiple channels for hotels worldwide. [Tech.eu]
Claira has secured US$7m in seed funding from major financial institutions to develop its AI-powered deal intelligence platform for investment analysis. [Business Wire]
Cluely has raised US$15m in Series A funding from Andreessen Horowitz, two months after securing US$5.3m in seed funding for its controversial AI-powered interview assistance tool. [TechCrunch]
Decagon has raised US$131m in Series C funding, reaching a US$1.5 billion valuation for its AI-powered customer service solutions used by companies like Hertz and Duolingo. [Yahoo Finance]
Delphi has raised US$16m in Series A funding to expand its AI platform that creates digital versions of experts for interactive, multilingual conversations. [FinSMEs]
Harvey AI has raised US$300m in Series E funding at a US$5 billion valuation to expand its legal AI solutions business serving law firms and corporate legal departments. [Fortune]
Jobright has raised US$3.2m in funding to enhance its AI-powered job search platform that automates applications and matches candidates with suitable positions. [FinSMEs]
Kognitos has launched a neurosymbolic AI platform for business automation with US$25m Series B funding, promising hallucination-free operations and comprehensive governance. [Business Wire]
Laude Institute has launched with US$100m from Databricks co-founder Andy Konwinski to fund independent AI research through grants and a nonprofit/commercial hybrid structure. [TechCrunch]
Lyceum has secured €10.3M in pre-seed funding to build sovereign European AI infrastructure, offering user-friendly GPU cloud services and EU-compliant data centers. [Tech.eu]
Mahalo has launched an AI-powered post-purchase platform and secured US$2.6m in pre-seed funding to help manufacturers streamline warranty management and customer support. [Business Wire]
Mary Technology has raised US$1.7m in pre-Seed funding for its Fact Management System, which transforms unstructured legal documents into searchable chronologies using LLMs. [Artificial Lawyer]
Maven AGI has raised US$50m in Series B funding to expand its enterprise AI platform for unifying customer journeys. [FinSMEs]
Metaview has secured US$35m in Series B funding to expand its AI-powered recruitment platform, which helps companies streamline hiring processes through automated note-taking and workflow optimization. [Yahoo Finance]
MiniMax has engaged advisers for a potential Hong Kong IPO that could value the Chinese AI company at US$3B, following its US$600 million funding round last year. [Verdict]
Norrsken Foundation has committed €300m to European AI startups focused on solving global challenges in climate, health, food, education, and society. [Tech.eu]
Pangram Labs has raised US$4m in seed funding to expand its AI-text detection technology used by businesses to assess product reviews. [FinSMEs]
Paraform has secured US$20m in Series A funding from Felicis to enhance its AI-powered hiring marketplace that empowers elite recruiters rather than replacing them. [Business Wire]
Profound has raised US$20m in Series A funding to expand its AI visibility platform, which helps marketers understand and control their presence in AI responses. [FinSMEs]
Quinn has raised US$11m in seed funding to expand its AI-powered financial planning platform that helps institutions deliver personalized advice at scale. [FinSMEs]
RevelAi Health has raised US$3.1m in seed funding to expand its conversational AI platform for musculoskeletal healthcare coordination and patient engagement services. [FinSMEs]
SuperDial has raised US$25m in Series A funding to advance its AI technology for automating insurance-related phone calls between healthcare providers and insurers. [FinSMEs]
Synthflow has raised US$20m Series A funding to expand its no-code voice AI platform, which has processed over 45m customer service calls since launching in 2023. [TechCrunch]
Thinking Machines Lab has raised nearly US$2B in seed funding led by Andreessen Horowitz for its AI collaboration venture, following Murati’s departure from OpenAI. [TechCentral]
Uncountable has raised US$27m in Series A funding to expand its cloud-based platform that helps industrial R&D teams move beyond spreadsheets. [Axios]
Wispr has raised US$30m in Series A funding to expand its AI voice app Flow, which converts speech to edited text across iOS and desktop platforms. [FinSMEs]
Acquisitions
Haveli Investments has agreed to acquire AI database company Couchbase for US$1.5B, offering shareholders US$24.50 per share. [Reuters]
Hebbia has acquired FlashDocs, expanding its AI platform’s capabilities to include automated presentation creation, serving financial, legal, and professional services firms. [Business Wire]
Rubrik has entered into an agreement to acquire Predibase, a developer platform for productionizing open source AI. [Business Wire]
There’s More
An analysis by Indeed revealed that companies’ use of ‘responsible AI’ terminology in job postings has been driven more by corporate image than regulatory compliance. [TechRadar]
LiveCodeBench Pro has launched a new benchmark system using algorithmic Olympiad problems to evaluate AI models’ coding abilities, revealing significant gaps between AI and human performance. [MIT Technology Review]
ServiceNow‘s CTO Pat Casey has reported that AI is automating junior engineers’ traditional coding tasks, forcing engineers to master new tools and specialize their roles. [Business Insider]
A survey by Tidio revealed that while 70% of users have sworn at AI chatbots, most remain polite and continue saying ‘thank you’ during interactions. [TechRadar]
xAI's Memphis data center has sparked controversy over air quality testing methods and results, with the SELC challenging the city’s assessment of pollution from natural gas turbines. [TechCrunch]
The Back Cover
Here’s The Hindley Street Country Club with a cover of Dire Straits’ Sultans of Swing.
If you know of a great cover version of a back-catalogue classic, drop an email to us at backcover@language-technology.com and we’ll consider it for inclusion here.
Got this from someone else?
Did you find this newsletter useful? if so, you might forward it to a friend; or you could email us at news@language-technology.com to tell us what you want more of.
Did you hate this newsletter? if so, you could forward it to an enemy; or you could email us at news@language-technology.com to tell us why – make it better!