Analyzing Daily Global News

When asked for a daily briefing, produce this structure immediately — no preamble, no "let me search first" narration in the final output:

DAILY GLOBAL BRIEFING — [Date]

1. TOP STORIES (3-6)
2. GENUINELY TRENDING (with noise-filter note)
3. UNDER-THE-RADAR / SMALLER COUNTRIES
4. BANGLADESH-RELEVANT DEVELOPMENTS
5. SOUTH ASIA & CONNECTED COUNTRIES WATCH
6. WATCHLIST (small stories that could become big)
7. VERIFICATION NOTES (contested/unconfirmed claims)

Each item gets: What happened (1-2 sentences, facts only) → Why it matters (impact reasoning) → Confidence (confirmed / developing / disputed) → Source count (how many independent outlets corroborate it).

Progress:

• Step 1: Establish "today" — determine current date/time and search only for last 24-48h developments; explicitly flag anything older being included for context
• Step 2: Broad sweep across categories (geopolitics, economics, conflict, climate, tech, health) using multiple independent sources per topic
• Step 3: Trend detection — separate high-volume social chatter from stories with real institutional/multi-source backing
• Step 4: Periphery scan — deliberately search smaller/developing countries and non-Western outlets, not just wire-service homepages
• Step 5: Bangladesh-relevance pass — run every major story through the impact filter (see below)
• Step 6: Regional deep-dive — India, Pakistan, Myanmar, China, Middle East, South Asia bloc
• Step 7: Cross-verify contested/high-stakes claims against 2-3 independent sources of different origin
• Step 8: Rank by impact, not coverage volume
• Step 9: Compose briefing with explicit confidence labels

Step 1: Establish "Today"

• Confirm current date. Treat anything not newly reported, updated, or materially escalated in the last 24-48h as background, not news.
• If a story has been running for days, only surface it if there's a genuine new development (new fact, escalation, reversal, official statement) — state explicitly what is new today vs. recap.

Step 2: Broad Sweep (Source Selection)

Query across source types, not just prominent outlets:

• Wire services (Reuters, AP, AFP) for baseline facts — highest reliability for "what happened"
• Regional/local outlets in the country of the event (e.g., Myanmar-based, Pakistani, Gulf outlets) for ground-level detail wires miss
• Financial/trade press (Bloomberg, Nikkei Asia, trade-specific outlets) for economic signals
• Official sources (government statements, central banks, UN/IMF/WHO releases) for primary confirmation
• Specialist/regional think tanks (e.g., South Asia-focused analysis) for context, clearly labeled as analysis, not fact

Avoid over-indexing on English-language Western media alone — it systematically under-covers the Global South stories this briefing needs.

Step 3: Trend Detection vs. Noise

A trend is "genuine" if it meets ≥2 of:

• Multiple independent, geographically diverse outlets report it within the window
• It has an identifiable causal event (not just a viral post/reaction cycle)
• Official or institutional actors are responding to it (statements, markets, policy)

Signs of noise/clickbait to exclude or downgrade:

• Single-source virality driven by outrage/emotion with no underlying event
• Celebrity/entertainment stories dominating engagement but with no broader consequence
• Headlines that outpace the actual verified content ("X slams Y" with no substantive claim inside)

Label explicitly: "Trending but low-substance — excluded from top stories" if something is loud but hollow.

Step 4: Periphery Scan (Smaller/Developing Countries)

Deliberately search for developments in: Sub-Saharan Africa, Central Asia, Southeast Asia (beyond Myanmar), Latin America, Pacific Islands, and smaller Middle Eastern/Gulf states. Ask: "What happened in a country I haven't checked yet?" Look for:

• Coups, elections, mass protests, sudden leadership changes
• Debt defaults, currency crises, IMF interventions
• Natural disasters, disease outbreaks
• New resource discoveries, major infrastructure/trade deals These often carry outsized global consequence (supply chains, migration, precedent-setting) despite thin coverage.

Step 5: Bangladesh-Relevance Impact Filter

For every candidate story, run this checklist — if any answer is "possibly," include it in the Bangladesh section with reasoning:

Step 6: Regional Deep-Dive (Priority Countries)

Always explicitly check for same-day developments in: India, Pakistan, Myanmar, China, Middle East/Gulf states, plus broader South Asia (Sri Lanka, Nepal, Bhutan, Maldives, Afghanistan). Even "no major development today" is worth a one-line note for continuity — but don't manufacture news where none exists.

Step 7: Cross-Verification

For anything contested, high-stakes, or with political incentive to distort:

• Require 2-3 independent sources of different national/ideological origin before stating as fact
• Explicitly separate confirmed fact (multiple corroborating primary/wire sources) / credible but unconfirmed (single strong source, or officials claim but not independently verified) / disputed (sources conflict) / speculation/analysis (pundit interpretation, clearly labeled as such)
• Note when a claim originates from an interested party (government press release, one side of a conflict) and hasn't been independently corroborated

Step 8: Prioritization Criteria (Impact > Volume)

Rank using these weighted factors, not "how many outlets covered it":

1. Scale of consequence — lives affected, economic magnitude, precedent set
2. Reversibility/urgency — is this a turning point or business-as-usual?
3. Second-order effects — could this cascade into markets, migration, conflict, alliances?
4. Bangladesh proximity — direct impact > indirect impact > no discernible impact
5. Novelty — genuinely new information today, not recycled coverage

A story with modest media coverage but high scale-of-consequence (e.g., a debt default in a small economy that could trigger IMF-wide contagion) outranks a heavily-covered but low-consequence story.

Step 9: Compose Output

Use the Quick Start structure. Keep each entry tight: fact, significance, confidence, source basis. Do not pad with filler transitions.

Example 1: Input: "Give me today's briefing." Output:

TOP STORIES
- [Country] central bank raises rates 75bps amid currency pressure — Confirmed (Reuters, Bloomberg, official statement). Why it matters: signals broader EM currency stress; watch for contagion to South Asian currencies.

GENUINELY TRENDING
- Viral claim about [event] — Trending but single-source origin, no institutional corroboration yet. Downgraded from top stories pending verification.

UNDER-THE-RADAR
- Mass protests in [small country] over fuel subsidy cuts — thin Western coverage, but risk of supply chain/migration spillover. Confirmed via local outlet + one wire report.

BANGLADESH-RELEVANT
- Red Sea shipping disruption escalates — Confirmed. Impact: longer transit times/higher freight costs for Bangladesh's EU-bound garment exports.

SOUTH ASIA WATCH
- India: [development]. Pakistan: no major new development today. Myanmar: border clashes near [region] — relevant to Rohingya situation, credible but not independently verified beyond one local source.

WATCHLIST
- Small dispute over [resource] in [region] — currently minor, but could escalate given [reasoning].

VERIFICATION NOTES
- Claim that [X] occurred is disputed: Source A confirms, Source B denies; treating as unconfirmed pending further reporting.

Example 2: Input: "Is there anything happening in Myanmar that matters for Bangladesh today?" Output: State only genuinely new developments (not recap of ongoing conflict background), confidence level, source basis, and specific link to Bangladesh (Rohingya repatriation, border security, refugee flows, trade). If nothing new, say so directly rather than restating old context as if new.