SoftBank Raises $11 Billion in Bonds to Fund OpenAI Stake

โดย Insider Monkey·JPUSGB·Read original
Summary · why it matters

SoftBank launched more than $11 billion of dollar and euro bonds on September 21 as it prepared for another $10 billion OpenAI installment. The financing sits alongside a broader credit line backed by Arm Holdings, which can reach about $25 billion against Arm shares, giving SoftBank liquidity without surrendering control of the chip designer. SoftBank also sold its entire roughly $5.8 billion stake in Nvidia in 2025, a move the article frames as a financing choice rather than a verdict on GPUs. Hedge fund positioning has shifted toward Arm, with 52 funds reporting longs in the second quarter of 2026 versus 46 in the prior quarter, while 285 funds held Nvidia, up from 275. The result is a hierarchy inside Masayoshi Son's AI portfolio: OpenAI is the cash sink he wants to fund, Arm is the strategic asset he wants to retain, and Nvidia was liquid enough to sell.

Impact on assets 5

Artificial Intelligence · 3 stocks
Financials · 1 stocks
SoftBank Group Corp.
9984
▲ PositiveCapitalrelevance

SoftBank raised over $11B in bonds and has an Arm-backed credit line up to ~$25B to fund its OpenAI stake

Semiconductors · 1 stocks

Theme Impact 3

Off-coverage companies 1

OpenAIPrivate▲ Positive
Capitalrelevance

SoftBank is preparing another $10B OpenAI installment funded by the new bond issuance

Related news

Executives of 3 AI Giants Sound Alarm at UN Security Council: "If It Becomes Uncontrollable, It Is an International Threat"

Executives from three of the world's leading artificial intelligence companies briefed the United Nations Security Council on the 23rd, warning that increasingly powerful AI systems could soon begin improving themselves and slip beyond human control, potentially becoming a threat to international security. At the meeting held alongside the annual UN General Assembly, three industry representatives spoke: OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, and Hugging Face co-founder Clement Delangue. Amodei said, "I even think that if it is not managed well, AI could become a risk to all of humanity," while Altman argued, "If AI is to be democratic, the most important decisions cannot be made only in a laboratory in San Francisco," stressing the need for international cooperation. Yoshua Bengio, co-chair of the UN's Independent International Scientific Panel on AI, expressed alarm, saying, "The dangers are real and imminent."
ロイター·1hRead more →
5impact 4

SoftBank Group Launches $11 Billion Junk Bond Sale to Fund OpenAI Investment

SoftBank Group launched an $11 billion junk bond sale to fund a major investment in OpenAI, a deal described as the largest non-financial corporate bond offering from the Asia Pacific and Japan region to date. The funding is a senior unsecured high yield issue of roughly $10 billion in U.S. dollars plus about €1 billion, targeting more than $11 billion in total. The planned $10 billion contribution to the third tranche of SoftBank's OpenAI stake is mostly debt funded, pointing to higher gross debt and interest costs until any offsetting asset sales or repayments occur. The first hard test will come in upcoming results, when SoftBank reports updated loan to value ratios, net debt, and interest expense after the bond closes, with a key marker being whether management pairs the new $11 billion of high yield funding with visible asset monetizations or repayments. SoftBank Group runs one of Japan's largest wireless telecom operations, and the move marks a push to extend that communications footprint into AI infrastructure and services tied to OpenAI's technology stack.
Simply Wall St·1hRead more →
impact 4

42 State AGs Build AI Agent Liability Framework as Congress Stalls

A bipartisan coalition of 42 state attorneys general is building the liability framework for agentic AI that Congress has yet to codify, filling a federal regulatory void confirmed by the Congressional Research Service. In December 2025, the coalition, led by officials from Pennsylvania, New Jersey, West Virginia, and Massachusetts, issued a coordinated letter to 13 major AI companies including Anthropic, Apple, Google, Meta, Microsoft, OpenAI, and xAI, demanding chatbot safeguards with a response deadline of January 16, 2026. State AGs are repurposing existing UDAP statutes, consumer protection laws, civil rights frameworks, and antitrust authority to police AI behavior, an approach that produced a first-of-its-kind settlement by Texas Attorney General Ken Paxton with healthcare AI company Pieces Technologies over false accuracy claims. Connecticut's AI Responsibility Act, signed in May 2026, grants the state AG exclusive enforcement authority under CUTPA starting October 1, 2026, with a mandatory one-year cure period through September 2027, while New Jersey's Fair Price Protection Act, effective August 2027, allows a private right of action for surveillance pricing with treble damages and no cure period. The Pennsylvania AG has sought preliminary injunctions against an AI company for falsely representing a chatbot as a licensed psychiatrist, and the Florida AG launched a criminal investigation in April 2026 after a violent incident involving a chatbot. State AGs filed seven antitrust actions in 2026 alone, surpassing the prior two years combined, including a 30-state coalition continuing the monopolization case against Live Nation and twelve states suing to block the Paramount-WBD merger despite federal declination.
Yahoo Finance·1hRead more →