April 30, 2026 — curated links and takeaways.
1. Anthropic releases Claude Mythos for cybersecurity testing
OpenAI and Anthropic both released frontier cybersecurity models; Anthropic's Claude Mythos Preview is being distributed to 11 organizations to identify vulnerabilities. This marks the first production deployment of frontier models for specialized security use cases, setting precedent for gated high-risk applications.
2. Meta launches Muse Spark from Superintelligence Labs
Meta released Muse Spark on April 8, 2026, the first in a new model series from Meta Superintelligence Labs. The launch signals Meta's shift toward foundational model development and direct competition with OpenAI and Anthropic at frontier scale.
3. Ineffable Intelligence raises $5.1B seed, Europe's largest AI funding to date
London-based Ineffable Intelligence closed the largest seed round in European AI history at $5.1B valuation, focusing on reinforcement learning. This indicates massive capital concentration in specialized AI capabilities and confirms Europe's emergence as a credible frontier AI funding destination.
4. Nautile-370M: efficient small language model for constrained inference
Researchers released Nautile-370M, a 371-parameter model optimized for reasoning under strict compute and memory budgets (arXiv:2604.24809v1). The sub-400M scale targets edge deployment and latency-critical applications where frontier models are unusable.
5. Netomi raises $110M Series C for AI customer service
Netomi closed a $110M Series C led by Accenture Ventures on April 30, 2026. The enterprise customer service category is consolidating around AI-native platforms with large-scale corporate backing.