We help you to learn,understand and update all the current affairs related to the business and startup world. We are not only a news channel but your guide to understand all complex economy, business, current affairs.
Join our Community
WhatsApp Group: chat.whatsapp.com/GigxBj9eE2FHZMnuxVsO9x
Telegram: t.me/+2Gzxhu_PJ2NhMzJl
#geopolitics #hotissue #recentnews #breakingnews #currentIssue #businessnews #rbi #sensex #business #voice #lapaasvoice
Lapaas Voice
₹10 a day is helping mutual fund investing travel beyond India’s metros.
PhonePe Mutual Funds says Daily SIP transactions grew nearly 5× between January and June 2026. The feature now has more than 500,000 users, with an average reported daily contribution of ₹50.
The geography is the bigger story: 77% of Daily SIP customers come from Tier-2 and Tier-3 cities. Around half are aged 18–29, showing how low-ticket, mobile-first investing is reaching a younger and wider audience.
But accessibility does not remove investment risk. Before starting any SIP, check the fund’s plan type, expense ratio, risk level and your time horizon. A daily SIP can support discipline; it cannot guarantee better returns.
7 hours ago | [YT] | 58
View 1 reply
Lapaas Voice
Your next mobile recharge could eventually ask a new question: 4G or 5G?
TRAI has proposed allowing telecom operators to offer separate tariff plans for the two network generations. If the government accepts the recommendation, 4G users could get plans tailored to value and compatibility, while 5G could become a clearer premium tier.
For operators, the split could create a new route to monetise years of 5G spectrum and infrastructure investment.
But nothing changes yet. The Department of Telecommunications can accept, modify or reject the recommendation. No new prices have been announced—and separate plans do not automatically guarantee cheaper 4G.
Swipe for the proposal, approval path and consumer impact.
11 hours ago | [YT] | 65
View 0 replies
Lapaas Voice
Meta has entered the terminal-coding race with Muse Code. 💻
The beta coding agent, powered by Muse Spark 1.2, is designed to work across large repositories—not just complete isolated lines. It can inspect a codebase, plan multi-file changes, write and debug code, run tests, validate results and coordinate persistent subagents.
Meta says one GPU-kernel optimisation evaluation involved more than 1,000 tool calls over as long as 24 hours. That is a company-reported stress test, not a guarantee for every software task.
The bigger shift: coding assistants are becoming software operators. Teams will need the same controls they use for human contributors—isolated branches, least-privilege access, tests, diff review and action logs.
11 hours ago | [YT] | 36
View 0 replies
Lapaas Voice
DeepSeek’s low-cost AI pitch is getting a time-of-day twist. ⏱️
A developer notice says the API will adopt peak–valley pricing, with peak-hour usage charged at 2× the regular rate across billing items. The activation date and exact time windows have not yet been published.
For teams building always-on apps and AI agents, that makes request timing part of cost engineering. The practical moves: map token usage by hour, add budget alerts, shift flexible batch jobs, use caching, and test fallback routes.
Important: this is an API pricing change. A broad consumer-chat price increase has not been confirmed.
14 hours ago | [YT] | 51
View 2 replies
Lapaas Voice
Central banks added a net 51 tonnes of gold to reported reserves in June 2026—up from 41 tonnes in May.
Poland and China led the month, while Uzbekistan and Kazakhstan remained active buyers. Across the first half, Poland was the largest reported buyer at 82 tonnes. Turkey and Russia were the biggest net sellers.
The bigger signal: gold remains strategic reserve ballast. It offers liquidity, diversification and no issuer credit risk, though it carries no yield and comes with price and storage trade-offs.
Swipe for the numbers and the reserve strategy behind them.
15 hours ago | [YT] | 54
View 0 replies
Lapaas Voice
Alakh Pandey has moved the Delhi High Court seeking protection against the unauthorised commercial use of his name, image, likeness, voice and identity online.
But this is not a blanket ban on memes or criticism.
The court has asked Pandey to sort the disputed content defendant by defendant—separating alleged commercial exploitation, vulgar or explicit misuse, and possible defamation. That distinction matters because satire, parody, fair criticism and legitimate commentary can still fall within protected speech.
The next step is the classification chart. After reviewing it, the court is expected to consider whether interim protection or targeted takedowns are justified.
The bigger question: how should India protect a public figure's digital identity without sanitising the internet?
Read the full story on LapaasVoice.com.
1 day ago | [YT] | 92
View 1 reply
Lapaas Voice
India’s delivery giants are playing the same market from different financial positions.
In Q1 FY27, Eternal reported ₹20,648 crore in adjusted revenue and ₹555 crore in adjusted EBITDA. Swiggy reported ₹7,112 crore in adjusted revenue and a ₹651 crore adjusted EBITDA loss—an improvement of ₹162 crore year-on-year.
The sharper contrast is in quick commerce:
• Blinkit: ₹17,132 crore NOV, +86% YoY, ₹102 crore adjusted EBITDA
• Instamart: ₹7,907 crore GOV, +39.8% YoY, ₹778 crore adjusted EBITDA loss
Instamart’s contribution margin improved to −0.2% for the quarter and reached break-even during May 2026. That is real progress—but contribution break-even is not the same as EBITDA profitability.
Eternal currently has greater financial flexibility. Swiggy’s case depends on turning improving unit economics into lower cash burn while sustaining growth.
Note: Blinkit reports NOV while Instamart reports GOV, and each company defines adjusted metrics in its disclosures. The figures are useful indicators, not perfectly like-for-like comparisons.
Read the full story on LapaasVoice.com.
1 day ago | [YT] | 61
View 0 replies
Lapaas Voice
Claude is bringing regional data controls closer to Indian enterprises—but “local processing” is not a single blanket setting.
Anthropic distinguishes between two controls: data residency, which determines where prompts, outputs and conversation history are stored; and inference residency, which determines where Claude processes a request and generates its response.
For regulated teams in banking, healthcare, government and telecom, that geographic control can remove a major blocker to AI adoption. But the fine print matters: availability depends on the enterprise deployment, cloud platform, endpoint, model and contract. It is not automatically enabled for every Claude account or every type of data.
The practical checklist: Where does the data rest? Where does inference run? Which connected tools or support processes can leave the region?
1 day ago | [YT] | 40
View 1 reply
Lapaas Voice
Long-running AI agents do not only need more context. They need the right memory to influence the right decision.
Meta AI researchers call the failure mode “behavioral state decay”: an agent identifies a requirement, failed command or open subgoal early in a task, but that information later stops shaping its actions—even when it still exists somewhere in the transcript.
Their proposed architecture adds a separate Memory Agent beside an unchanged Action Agent.
At fixed intervals, the Memory Agent:
• Reviews recent trajectory steps
• Updates a structured memory bank
• Tracks stable facts, execution history and unresolved risks
• Decides whether to inject a targeted reminder—or stay silent
With Claude Opus 4.6 as the Memory Agent and Sonnet 4.5 as the Action Agent, Meta reports pass@1 gains of 8.3 percentage points on Terminal-Bench 2.0 and 6.8 points on τ²-Bench.
The important idea is selective intervention. Too little memory allows repeated mistakes; too much adds tokens, latency and distraction.
This is a research architecture, not a released Meta AI consumer feature.
1 day ago | [YT] | 44
View 0 replies
Lapaas Voice
UPI may not remain zero-MDR for every merchant transaction forever.
The Centre is reportedly considering a 0.25%–0.35% merchant discount rate on eligible high-value UPI payments received by large businesses. The proposed fee would be deducted from the merchant—not charged directly to the customer—and consumers, person-to-person transfers and smaller merchants are expected to remain protected under the models being discussed.
At the reported range:
• ₹5,000 payment → ₹12.50–₹17.50 MDR
• ₹10,000 payment → ₹25–₹35 MDR
• ₹1,00,000 payment → ₹250–₹350 MDR
Why target larger payments? Jefferies estimates cited by Lapaas Voice suggest transactions above ₹2,000 form only about 4% of merchant-payment volume but around 67% of value.
Important: no final rate, transaction threshold, merchant-turnover cut-off or start date has been notified. Public reports have cited different possible rate structures.
Read the full story on LapaasVoice.com.
1 day ago | [YT] | 47
View 5 replies
Load more