Try Artisan’s All-in-one Outbound Sales Platform & AI BDR

Ava automates your entire outbound demand generation so you can get leads delivered to your inbox on autopilot. She operates within the Artisan platform, which consolidates every tool you need for outbound:

• 300M+ High-Quality B2B Prospects, including E-Commerce and Local Business Leads

• Automated Lead Enrichment With 10+ Data Sources

• Full Email Deliverability Management

• Multi-Channel Outreach Across Email & LinkedIn

• Human-Level Personalization

Book a demo to see what Ava can do.

Today:

• Claude now includes web search

• OpenAI launches new audio models

• NotebookLM adds Mind Maps & more

• SoftBank acquires Ampere for $6.5B

• Perplexity AI raising $1B

Claude now includes web search

Claude now includes web search, enabling real-time access to the latest information. This feature improves accuracy and relevance for users needing up-to-date insights. It also provides source citations for transparency. Sales teams, financial analysts, researchers, and shoppers can benefit from better-informed decisions. Currently available for paid U.S. users, the feature will expand soon. Users can enable it in settings and start using Claude 3.7 Sonnet for web-informed responses.

Why It Matters

1. Real-Time Knowledge – AI models often rely on outdated training data; web search ensures fresh, relevant information.

2. Improved Accuracy – Source citations help reduce misinformation and enhance trust in AI-generated content.

3. Expanded Use Cases – Businesses, researchers, and individuals can use AI for more informed decision-making.

OpenAI launches new audio models

OpenAI has launched new speech-to-text and text-to-speech models in its API, improving accuracy and customization for voice agents. The speech-to-text models reduce errors, even in noisy settings, while the text-to-speech model allows developers to control speaking styles. These advancements enhance applications like customer service and meeting transcription. Available now for developers, OpenAI aims to further refine its audio models and expand into multimodal experiences, including video integration.

Why It Matters

1. More Natural AI Conversations – Customizable speech improves human-like interactions.

2. Higher Transcription Accuracy – Reduces errors in diverse languages and environments.

3. Advancing Multimodal AI – Sets the stage for integrating voice, text, and video capabilities.

NotebookLM adds Mind Maps & more

Google’s NotebookLM now includes Mind Maps for visualizing complex topics and an output selector for generating text in different languages. These features enhance productivity for students, professionals, and researchers by improving organization and multilingual collaboration. Users can now better structure information and work in their preferred language. The updates will gradually roll out to Google Workspace and NotebookLM Plus users over the next 15 days.

Why It Matters

1. Enhanced User Interaction – AI can now help users structure and visualize information, making research more intuitive.

2. Multilingual AI Capabilities – Expanding language support improves accessibility and global collaboration.

3. Boosts AI Adoption – More professionals and students may rely on AI-powered tools for productivity.

🧠RESEARCH

φ-Decoding: Adaptive Foresight Sampling for Balanced Inference-Time Exploration and Exploitation

phi-Decoding is a method to improve AI text generation by balancing exploration (trying new possibilities) and exploitation (choosing the best option). It predicts future steps for better decision-making and uses pruning techniques to boost efficiency. Tested across benchmarks, it outperforms existing methods and works well with various AI models.

DeepMesh: Auto-Regressive Artist-mesh Creation with Reinforcement Learning

DeepMesh improves 3D model creation by enhancing structure and detail in triangle meshes. It introduces better pre-training with a new tokenization method and uses Reinforcement Learning to align with human preferences. By refining data processing and scoring quality, it generates more complete and visually accurate meshes, outperforming existing approaches.

TULIP: Towards Unified Language-Image Pretraining

TULIP enhances AI models that understand both images and text. Unlike existing models that prioritize language over fine visual details, it improves image comprehension using better data augmentation and contrastive learning. Scaling to over 1 billion parameters, it sets new performance records in image classification and vision-language tasks.

🛠️TOP TOOLS

Neural Love - AI-powered platform offering free image generation, enhancement, and media processing tools.

Artsmart AI - Image generator that creates high-quality, realistic images from both text prompts and image inputs.

Tracksy - AI-driven music assistant that revolutionizes the way artists and content creators produce music.

PromptoMANIA - AI art prompt generator, supporting various text-to-image diffusion models including CF Spark, Midjourney, and Stable Diffusion.

Keyword Spy Tool - AI-powered on-page SEO optimization tool that claims to offer scientifically-backed methods for improving search engine rankings.

📲SOCIAL MEDIA

🗞️MORE NEWS

• ​SoftBank acquires chip start-up Ampere Computing for $6.5 billion, enhancing its AI infrastructure strategy. Ampere, founded by ex-Intel president Renée James, designs processors for cloud computing using Arm technology. The acquisition aligns with SoftBank's AI goals, complementing its earlier purchase of Graphcore. ​

• Perplexity AI, backed by Nvidia and Jeff Bezos, is raising up to $1 billion at an $18 billion valuation. The AI search startup competes with Google and OpenAI, expanding its platform despite plagiarism accusations from media organizations.

• The Aardvark AI project aims to make advanced weather forecasting accessible via desktop computers, benefiting developing nations. Led by the UK’s Alan Turing Institute, it provides fast, energy-efficient predictions using satellite and sensor data.

• Meta AI is expanding across 41 European countries, integrating into Facebook, Instagram, WhatsApp, and Messenger. Users can access AI-powered chat in six languages, enhancing group chats and content discovery.