AI Practical Use #3#: Let AI help you with Excel data analysis. AI 实用玩法第 3 个： 让 AI 帮你做 Excel 数据分析。 Here is a very common office situation: You have an Excel file with sales data, costs, profit, regions, products, and dates. Normally, you may spend 2 hours writing formulas, checking data, making summaries, and building charts. But with AI, you can finish the first draft in about 10 minutes. 一个很常见的办公场景： 你手里有一份 Excel 数据， 里面有销售额、成本、利润、区域、产品、日期。 以前你可能要花 2 小时： 写公式、查数据、做汇总、看趋势、做图表。 现在可以先交给 AI， 10 分钟生成初步分析结果。 You don’t need to manually type every complex formula. Let AI help you: Build formulas Summarize key findings Find abnormal data Compare trends Suggest chart formats Create a report structure 你不需要自己一个个输入复杂函数。 可以让 AI 帮你： 生成公式 总结关键结论 找出异常数据 对比趋势变化 建议图表形式 生成汇报框架 Here is a simple prompt: 这里有一个简单提示词： Please analyze this Excel data. Help me build the right formulas, summarize the key findings, find possible errors or abnormal values, and suggest the best chart or report format. I will review and verify the final results. 中文版本： 请分析这份 Excel 数据。 帮我生成合适的公式，总结关键结论，找出可能的错误或异常值，并建议最适合的图表或汇报格式。 最终结果由我来审核确认。 The key idea is simple: AI does the heavy first draft. You review the logic and final result. 核心思路很简单： AI 负责先把复杂工作做出来， 你负责审核逻辑和最终结果。 Before: 2 hours manually writing formulas. After: 10 minutes with AI assistance. 以前： 手动写公式、做分析，可能要 2 小时。 现在： 借助 AI，10 分钟先完成初稿。 AI is not here to replace your judgment. It helps you save time on repetitive work, so you can focus on checking, thinking, and making better decisions. AI 不是替代你的判断力。 它是帮你节省重复劳动的时间， 让你把精力放在审核、思考和决策上。 Let AI write the formulas. You review the results. 让 AI 写公式， 你负责审核结果。 That is a smarter way to work. 这才是更聪明的办公方式。 #ChatGPT# #AI# #AITools# #Excel# #ExcelTips# #DataAnalysis# #Productivity# #WorkSmarter# #OfficeWork# #BusinessTools# #Automation# #DigitalTools# #TechTips# #FutureOfWork# #PromptEngineering#

Frédéric Renken and Steijn Pelle during the year they spent doing manual back office work at doctors' offices, before starting @LassieAI to automate it Full profile:

Grok 4.7 just outperformed GPT-6 Astra on professional work On the same Artificial Analysis run: • GDPval-AA — Grok leads by +153 Elo • AA-Briefcase — Grok leads by +88 Elo Grok 4.7 is now ahead on both professional deliverables and multi-hour office work SpaceXAI is pushing Grok hard into real-world knowledge work

BREAKING: Grok 4.7 beats GPT-6 Astra in both real-world work benchmarks: • Professional knowledge work: 1,695 vs 1,542 • Multi-hour office work: 1,657 vs 1,569 Grok 4.7 is the clear winner across both. 🔥

Beyond the Model: Training a 397B Agent with SkyRL | Charlie Ruan (阮世麟) What does it take to make RL work at 397B scale—beyond training the model itself? Join Charlie Ruan (@charlie_ruan) for a deep dive into SkyRL and the hard-won lessons of training a 397B office-work agent—from fixing environment and harness bugs to system tuning and the final hero run. Hosted by @qingke_ai and simulcast on Zhihu Frontier. 📅 Sept 19 · 10 AM Beijing / 2 AM UTC 📺 Watch live:

✈️ @SouthwestAir started with an FAQ agent Then customers asked: actually… can you just do this for me? Agent Script takes it from answers to action: → AI Assistant handles the conversation → Routes back-office work to specialized agents → Keeps business rules fixed where they need to be + leaves room to reason where they don’t Action beats answers every time

Sam Altman is probably shitting his pants today... Grok was an early beta in November 2023. And today, xAI shipped Grok 4.7, in just under 3 yrs. ChatGPT had already been live for about 11 months before Grok even existed... OpenAI had the head start. On the 7 hard-work tests xAI published today, Grok 4.7 beat OpenAI's GPT-5.6 Sol on 5 of them. 1/ Coding, the longer jobs... Grok 46.3%. Sol 41.7%. 2/ Electrical engineering, actually designing circuits... Grok 64%. Sol 39.4%. 3/ Multi-hour office work... Grok 1,657. Sol 1,487. 4/ Multi-hour work in the computer terminal... Grok 38.0%, Sol 37.3%. 5/ Legal work... Grok 19.6%. Sol 2.5%. Sol still leads the other 2. 6/ A second coding test... Sol 72.7%. Grok 71%. 7/ Doctor-style reasoning... Sol 60.5%. Grok 56.7%. Then an outside lab (Artificial Analysis) graded them... Grok got a 46. Sol got a 47. Just a one point difference. On coding agents, meaning an AI that writes the code and then runs it, that same lab had Grok at 56 and Sol at 55... Grok passed Sol. FYI, GPT-6 and Claude are still at 62. And one more from that same lab, bc this is the test that looks like a real job. The AI had to hand you the document, the spreadsheet, the slides... Grok scored 1,695. Sol scored 1,588. OpenAI's newer GPT-6 scored 1,542. Claude was still first, at 1,735. And FYI, the price did not go up... Grok 4.7 still costs the same as Grok 4.6 with $2 per million tokens in, $6 out (Sol is $4 in and $20 out). Materially more capabilities, but Grok is still the same price... It's awesome to see Grok caught up to a company that was already about a year ahead, in just 2 years and 10 months. And I bet soon, Grok will surpass them all... Elon is back with SpaceXAI!

There’s a ton of “if you had invested $1,000 in 25 years ago, you’d have x today!”, but sometimes it’s interesting to look at the rest of them. What if you had invested in one of the most exciting growth names a century ago? In 1926, the major market themes were aviation, office/industrial automation and electrification. One of the best performing big board stocks over the next 3 years was Burroughs Adding Machines - a company that made, you guessed it, adding machines. It rode the proliferation of office work to a +1,996% gain over the next 168 weeks. And it still survives today. It acquired Sperry Rand in 1986 and the combined company was renamed Unisys, which trades today as UIS. If you had invested $1,000 in Burroughs Adding Machines in 1926, the shares would have been worth ~$20,900 at the 1929 peak. And $2,800 today. But Burroughs, like many companies a century ago, was an aggressive dividend payer. Assuming you reinvested dividends, you’d have around $35,000 today. Seems decent, you outperformed the consumer price index by about 85%. But you’ve underperformed the index by a factor of 640 to 1. That same $1,000 invested in the S&P500 (or similar large cap U.S. index, using Shiller/BLS data) would be worth $22.4 million today. An investor in 1926 picks a defining technology winner of their generation, a company that manages to continue to exist for a century. But a century was long enough to go through disruption -> hypergrowth -> dominance -> maturity -> tech transition -> renewed relevance -> consolidation -> new disruption -> terminal decline. I don’t know exactly what the point is here but I find it pretty interesting considering most of the historical stock selection focuses on companies that survived and thrived.