WEEKLY DIGEST
Week in AI: Jun 29 — Jul 03, 2026
The top stories from our coverage this week.
@Google Research's Gemini-SQL2 tops text-to-SQL benchmarks at 80.04% accuracy, outperforming OpenAI and Anthropic.
https://huggingface.co/papers/2606.27277
J.P. Morgan warns of investor exuberance as 42 AI companies in the S&P 500 account for 65-80% of the index's profits.
https://the-decoder.com/j-p-morgan-sees-a-pile-of-red-flags-in-the-ai-market/
@Meta AI releases Brain2Qwerty v2, a non-invasive MEG brain-to-text pipeline decoding typed sentences at 61% word accuracy.
https://www.marktechpost.com/2026/06/30/meta-ai-releases-brain2qwerty-v2-a-non-invasive-meg-brain-to-text-pipeline-decoding-typed-sentences-at-61-word-accuracy/
Meta is building a cloud business to sell spare AI compute to outside customers, following SpaceX's playbook.
https://the-decoder.com/meta-follows-spacexs-playbook-and-builds-a-cloud-business-to-sell-its-spare-ai-compute-to-outside-customers/
Microsoft launches $2.5B Frontier Company to embed 6,000 AI engineers directly into enterprise clients.
https://the-decoder.com/microsoft-launches-2-5-billion-frontier-company-to-embed-6000-ai-engineers-inside-enterprise-clients/
Only three AI models survived 500 days running a simulated software company without going broke.
https://the-decoder.com/only-three-ai-models-finished-above-starting-capital-in-a-500-day-startup-survival-test/
Meta restricts engineers' use of @AnthropicAI's Claude and OpenAI's Codex to keep rival AI out of its training data.
https://the-decoder.com/meta-restricts-use-of-claude-code-and-codex-to-keep-rival-ai-out-of-its-training-data/
Queue raises funding to build fully autonomous pharmacies for hospitals and underserved communities.
https://www.therobotreport.com/queue-raises-funding-fully-autonomous-pharmacy/
Anthropic reportedly in talks with Samsung to manufacture custom AI chips while keeping Nvidia close.
https://the-decoder.com/anthropic-reportedly-explores-custom-chip-manufacturing-with-samsung-while-insisting-nvidia-still-matters/
@DeepSeek_AI released DSpark, a speculative decoding framework that accelerates DeepSeek-V4 per-user generation by 60–85% over MTP-1.
https://www.marktechpost.com/2026/06/27/deepseek-releases-dspark-a-speculative-decoding-framework-that-accelerates-deepseek-v4-per-user-generation-60-85-over-mtp-1/