SoftBank has launched one of the largest corporate junk bond sales ever, aiming to raise over $11.1 billion to fund its aggressive push into artificial intelligence. The deal includes a $10 billion dollar portion and a €1 billion euro component, with bond yields reaching as high as 9.75 percent. CEO Masayoshi Son wants to position SoftBank as the world's leading AI platform, and the proceeds will flow into the semiconductor supply chain, data centers, and large language model development.
Key Takeaways
- SoftBank raised a record $11.1 billion through a high-yield corporate bond sale, the largest globally on record, to fund AI investments
- SoftBank has committed $64.6 billion to OpenAI and will own about 13 percent of the company
- SoftBank's investments span Intel, Nvidia, Arm Holdings, Graphcore, and ABB's robotics business, with its Arm stake worth over $232 billion
- Bond yields of up to 9.75 percent are much higher than a similar 2021 issuance, reflecting growing investor caution about AI-related debt
- Illinois and Oregon are both advancing AI regulation, with Illinois creating an AI Cabinet through 2027 and Oregon ordering new safety rules within 90 days
- Reddit won a court ruling against Anthropic over AI training data, strengthening Reddit's control over its content and supporting paid data licensing plans
- Meta's AI agent app Muse reached 2.5 million downloads since its September 8 debut, outpacing rivals like OpenAI and Anthropic in consumer-focused AI
- A survey found 14 percent of job seekers would accept lower pay for AI training, and 42 percent say AI training offerings make a company more attractive
- Okta warns that AI agents create new insider security risks, with one financial firm finding about 13,000 agents but only 1,000 valid
- The EU proposed AI standards for ship navigation to the International Maritime Organization, targeting safety-critical AI systems with a goal of completion by 2029
SoftBank launches massive junk bond sale to fund AI investments
SoftBank Group started one of the largest corporate junk bond sales ever, aiming to raise over $11 billion. The money will fund large-scale investments in artificial intelligence. The deal includes a $10 billion dollar part and a €1 billion euro portion. This adds to SoftBank's growing debt-funded AI strategy.
SoftBank's major AI and semiconductor investments detailed
SoftBank founder Masayoshi Son is building SoftBank into the world's leading AI platform. The company has invested heavily in OpenAI, Intel, Arm Holdings, Nvidia, and other AI and semiconductor companies. SoftBank's stake in Arm is worth over $232 billion. The company also acquired Graphcore and is buying ABB's robotics business.
SoftBank issues $11.1 billion in bonds for OpenAI funding
SoftBank issued $11.1 billion in bonds with interest rates as high as 9.75% to fund its OpenAI investments. The deal is the largest high-yield corporate bond issuance globally on record. SoftBank has committed $64.6 billion to OpenAI and will own about 13% of the company. The bond yields are much higher than a similar 2021 issuance.
SoftBank plans record $11.1 billion high-yield bond sale
SoftBank Group announced plans to raise a record $11.1 billion through low-rated corporate bonds. The bonds offer yields of up to 9.75% to attract investors. The sale includes both dollar- and euro-denominated notes. This comes as caution grows among investors about AI-related debt.
SoftBank raises record $11.1 billion for AI investments
SoftBank raised $11.1 billion in the largest high-yield corporate bond sale globally on record. CEO Masayoshi Son aims to make SoftBank the world's leading AI platform. The funds will support investments across the semiconductor supply chain, data centers, and large language model development. The company has made major bets on OpenAI, Arm, and robotics.
PGA of America keeps AI simple with strong identity security
The PGA of America uses a lean technology team and a simplified approach to manage AI across its systems. It guides employees toward a common AI platform and uses identity controls around tools like MCP and agent gateways. The organization also brings finance, HR and legal teams into technology decisions. Scott said the approach reduces complexity and makes AI adoption manageable.
Okta highlights AI agent risks and the need for recovery plans
Okta says AI agents create a new insider security risk because they can act across apps and delegate tasks to other agents. The company calls for discovery, authorization and runtime controls to manage these agents. A financial firm found about 13,000 agents in its environment but only 1,000 were valid. Okta also stresses the need for clear human ownership and limits on agent permissions.
Illinois creates AI cabinet to study regulation and safety
Illinois Governor JB Pritzker signed an executive order to create an AI Cabinet made up of state agency representatives. The unpaid group will develop policy proposals through 2027 to assess risks to residents and critical infrastructure. It will also review data center energy use, procurement practices and legal accountability for AI companies. Illinois already passed a law requiring annual AI audits starting in 2028.
Oregon governor orders new AI safety rules
Oregon Governor Tina Kotek issued an executive order requiring state officials to develop rules for advanced AI. The order calls for qualifications for third-party reviewers and a definition for frontier AI models. It also asks whether a kill switch for unsafe AI systems is possible. The state's chief information officer has 90 days to propose regulations, citing a lack of federal action.
Reddit court win over AI data use may affect stock outlook
Reddit won a court ruling against Anthropic over how AI training data is used. The judge agreed that Anthropic must follow Reddit's user agreement. This strengthens Reddit's control over its data and supports its plans for paid data licensing. Analysts still focus on ad growth and user trends, but the ruling adds a new layer to Reddit's investment story.
First steps to secure AI agent data in 90 days
Kelly Herrell, CEO at Ticketing, advises organizations to focus on visibility in week one. Map the paths between agents, data sources, models, and tools. Measure sensitive data crossing those boundaries and enforce policy where data moves. Postpone big projects like redesigning identity systems or rewriting every agent until later.
Meta's Muse app gains 2.5 million downloads
Meta's AI agent app Muse has reached 2.5 million downloads since its September 8 debut. The app is designed for everyday users rather than complex business tasks. Rivals like OpenAI and Anthropic did not match Meta's consumer-focused approach. Muse is the first publicly available AI agent built by a major tech company.
Job seekers accept lower pay for AI training
A survey of 1,000 U.S. adults by ICIMS found that 14% of job seekers would take lower pay for AI training. About 42% say a company offering AI training is more attractive. Nearly half of U.S. adults have built AI skills in the past six months. Employers are increasingly paying a premium for AI skills.
Morgan Freeman calls AI voice replication stealing
Actor Morgan Freeman, 89, said using AI to copy a person's voice is stealing. He advised affected people to seek legal help. Freeman acknowledged the difficulty of tracing where AI-generated content originates. He also spoke about his work to preserve blues music and his involvement with blues club Ground Zero in Mississippi.
EU proposes AI standards for ship navigation
The European Union proposed AI standards for navigation and communication equipment on ships to the International Maritime Organization. The proposal targets AI systems that perform safety functions on vessels. The IMO's Maritime Safety Committee will review it in London in December 2026. The EU wants standards completed by 2029.
Cerebras and SoundHound compete as top AI stocks for 2026
Cerebras and SoundHound are two AI tech stocks drawing investor attention. Cerebras reported 76% revenue growth and swung to profitability. SoundHound is narrowing its losses by 20% while integrating major acquisitions. The better buy depends on investment goals and risk tolerance.
Sources
- SoftBank Starts Jumbo High-Yield Bond Sale to Finance AI Push
- SoftBank's growing bets on AI and semiconductor assets
- SoftBank issues $11.1 billion in bonds in OpenAI financing push
- SoftBank Group plans record sales of high-yield bonds to invest in OpenAI
- Factbox-SoftBank's growing bets on AI and semiconductor assets
- Identity security shapes the PGA of America’s AI strategy
- AI agent security goes beyond access
- Illinois Governor JB Pritzker establishes AI cabinet amid calls for greater regulation
- More AI regulation ordered by Oregon governor
- Court Win Over AI Data Use Might Change The Case For Investing In Reddit Stock (RDDT)
- What to do first when you get 90 days to secure AI agent data
- Meta Keeps Muse Momentum as Rivals Push New AI Models
- Job Seekers Willing to Take Lower Pay to Get AI Training
- Morgan Freeman: Using artificial intelligence to replicate voice is ‘stealing’
- The EU proposes to the IMO specific standards for artificial intelligence in navigation and communications equipment
- Cerebras vs. SoundHound: Which AI Tech Stock Is a Better Buy in 2026?
Comments
Please log in to post a comment.