Loading / 加载中
Thinkgap feed
A New York Times Magazine article examines Larry Ellison's aggressive investments in artificial intelligence through Oracle and questions whether he risks becoming the most visible symbol of an AI bubble. The piece explores the scale of his bets and the potential consequences as AI hype intensifies.
ti is a terminal-based AI agent for quantitative analysis and trading, similar to Claude Code but tailored for quants and traders. It supports natural language backtesting—for example, "backtest AMD vs NVDA past two years"—automatically fetching data and running the analysis. The tool includes a built-in mean-variance portfolio optimization engine and provides proprietary data on US Congress members' and corporate insiders' stock trades (240+ Congress members, 10,800+ insiders). Real-time market quotes, news, and research reports are also integrated. Users can install via `bun i -g @tradeinsight-info/ti` and access a free trial after registration.
OpenCode Go now supports DeepSeek's official version, but users must enable a background authorization to avoid errors. The quota and pricing remain unchanged. In real-world tests, the flash official version runs previously pro-level tasks without issues and is significantly faster. The author speculates that if the flash model surpasses GLM5.2 in benchmarks, the $10/month OpenCode Go subscription would become extremely cost-effective.
This item links to an analysis page on Artificial Analysis for the DeepSeek V4 Flash 0731 model, covering intelligence, performance, and pricing. The provided raw content is limited to the page title, containing no specific metrics, benchmark results, or cost figures. No further details about the model's capabilities or market positioning are available from this snippet.
OpenAI cut GPT-5.6 Luna pricing by 80% to $0.20 per million input tokens, and Terra by 20% to $2 per million. DeepSeek is reportedly planning a 1GW AI data center in Ulanqab, Inner Mongolia, with phased operation expected by late 2027 or early 2028. Anthropic's Texas AI campus will be backed by Google, involving $15 billion in financing from Morgan Stanley for a 1.6GW data center and natural‑gas plant; Google guarantees lease obligations for a ~20% equity stake, and the site will run on Google TPUs. The EU opened an AI gigafactory tender offering up to €10 billion in public funding for up to seven projects, each with at least 100,000 advanced AI chips. Amazon raised its 2026 cash CapEx to $220 billion due to surging AI demand and higher memory costs, yet still expects a capacity shortfall into 2027.
Ethan Mollick warns that an economic downturn in the next few years could push companies to deploy AI primarily for cost-cutting rather than expanding the role of humans. He views this as a bad precedent. He emphasizes that R&D investment is needed to enable people to use AI in more meaningful ways, but that requires continued spending.